They have no idea

[Lance Berg]

From the SOE page on updates
http://eqplayers.station.sony.com/game_updates.vm
:

We currently do not have information about the next Game Update.

All Updates




Upcoming Upcoming Updates
7/25/2007 July 25, 2007

Its 7/31, they've had this update scheduled since at least last night
because I heard people talking about it in general channel. And yet a)
they have no information about it and b) don't even know its going on!

Patch had been going for 30 minutes when I checked this.

[Eric D. Braden]

From the SOE page on updateshttp://eqplayers.station.sony.com/game_updates.vm
We currently do not have information about the next Game Update.
All Updates
Upcoming Upcoming Updates
7/25/2007 July 25, 2007
Its 7/31, they've had this update scheduled since at least last night
because I heard people talking about it in general channel. And yet a)
they have no information about it and b) don't even know its going on!
Patch had been going for 30 minutes when I checked this.

I was on about the same time (6:30 CST). I didn't check the official
"updates" page, but it was on the Live forums.

[Tony Evans]

Post by Eric D. Braden
I was on about the same time (6:30 CST). I didn't check the official
"updates" page, but it was on the Live forums.

It's been on the live forums every time, virtually 99% correct, for the
past 24 months. They usually update if the patch is over running, and when
the servers are up, and any issues, in the little panel Zat has setup.

Lance refuses to read the forums and refuses to help himself, and instead
insists on using a resource we all know is inaccurate.

And then moaning, every, single, time.

[the wharf rat]

Post by Tony Evans
And then moaning, every, single, time.

It's common courtesy to put system downtimes in MOTD. There's
no reason I should log in at 5am and find out that the servers are going
down in 3 minutes. I shouldn't have to google for forums to find out
when a service I pay for is having yet another emergency downtime. Why
can they put the damn fan faire conventions in MOTD but can't put a notice
of downtime in even 2 hours beforehand?

This is the same lack of professionalism on the part of the EQ
operations team that leads to situations like "We changed all the AA code
but didn't check to see if it really worked".

And putting up network status pages and updating them AFTER the
fact is just stupid.

[Tony Evans]

Post by the wharf rat
It's common courtesy to put system downtimes in MOTD. There's
no reason I should log in at 5am and find out that the servers are going
down in 3 minutes. I shouldn't have to google for forums to find out
when a service I pay for is having yet another emergency downtime. Why
can they put the damn fan faire conventions in MOTD but can't put a notice
of downtime in even 2 hours beforehand?
This is the same lack of professionalism on the part of the EQ
operations team that leads to situations like "We changed all the AA code
but didn't check to see if it really worked".
And putting up network status pages and updating them AFTER the
fact is just stupid.

It's all true.

It's all irrelevant to the issue.

They DO post the information in an EASY to read place on the FRONT page of
the Forums *usually*. It's very useful, it's very easy to check.

It's usually updated during the patch with relevant information.

Not checking it, and then moaning about not knowing what is going on is
futile.

[the wharf rat]

Post by Tony Evans
They DO post the information in an EASY to read place on the FRONT page of
the Forums *usually*. It's very useful, it's very easy to check.

And you know this how?

Because the network status page includes directions in 24 point
type instructing you to stop expecting the network status page to contain
network status and go find The Forums? (Which Forum did you say it was
by the way? "Test", or something equally intuitive, wasn't it?)

Post by Tony Evans
Not checking it, and then moaning about not knowing what is going on is
futile.

Yes, I'm familiar with this attitude. "It was difficult to
design so it should be difficult to use". How quaint to cling to such
thinking in this modern era. Sony and its apologists: coelocanths sculling
across the slimy bottom of the sea of service...


"Oh no, we won't give in
we'll go livin'
in the past..."

[Tony Evans]

Post by the wharf rat
Because the network status page includes directions in 24 point
type instructing you to stop expecting the network status page to contain
network status and go find The Forums? (Which Forum did you say it was
by the way? "Test", or something equally intuitive, wasn't it?)

No, it's THE Everquest Forums.

The front page of them.

Start with EQPlayers.com (if you must), click Forums, read the text in the
box which at the moment says "The EverQuest Servers are up!~ Zatozia" but
early gave information on the servers being down.

Not exactly hard to find.

Post by the wharf rat
Yes, I'm familiar with this attitude. "It was difficult to
design so it should be difficult to use". How quaint to cling to such
thinking in this modern era. Sony and its apologists: coelocanths sculling
across the slimy bottom of the sea of service...

Huh? I agreed that the eqplayers stuff sucks, I'm saying there IS an option
that works, which is EASY to use and find, and that by NOT using it and
then complaining about the one everyone knows doesn't work Lance is being
intentionally obtuse.

[Lance Berg]

Post by Tony Evans

Post by the wharf rat
Because the network status page includes directions in 24 point
type instructing you to stop expecting the network status page to contain
network status and go find The Forums? (Which Forum did you say it was
by the way? "Test", or something equally intuitive, wasn't it?)

No, it's THE Everquest Forums.
The front page of them.
Start with EQPlayers.com (if you must), click Forums, read the text in the
box which at the moment says "The EverQuest Servers are up!~ Zatozia" but
early gave information on the servers being down.
Not exactly hard to find.

Post by the wharf rat
Yes, I'm familiar with this attitude. "It was difficult to
design so it should be difficult to use". How quaint to cling to such
thinking in this modern era. Sony and its apologists: coelocanths sculling
across the slimy bottom of the sea of service...

Huh? I agreed that the eqplayers stuff sucks, I'm saying there IS an option
that works, which is EASY to use and find, and that by NOT using it and
then complaining about the one everyone knows doesn't work Lance is being
intentionally obtuse.

I'm not being intentionally obtuse. There's a resource, which would be
perfect for communicating the simple facts of the matter. Here's our
next update, when it will start, when we guess it will stop. Hey, we've
been working on it for a while and it looks like it will indeed be
finished on time/go over time maybe an hour longer than we thought/be
done early.

Here's what we are trying to change.

Its right there, in black and white, they update it so its not a dead page.

BUT its almost never accurate. Three minutes of someone's day.

I spend more time than that complaining about it here on AGE :)

Post a link to this forum you claim has the accurate poop, and I'll
bookmark it and try looking there.

BUT if the EQplayer page doesn't match that, I'm still gonna complain.

If you can't keep the page up to date, then just take it down, there's
no point at all in the page as it is. If minimal attention were paid to
keeping it up to date it would be superior to wandering around on some
web forum. Right now its just misleading and irritating.

[Tony Evans]

Post by Lance Berg
Post a link to this forum you claim has the accurate poop, and I'll
bookmark it and try looking there.

Click 'Forums' on the EQPlayers main page you love so much.

Post by Lance Berg
BUT if the EQplayer page doesn't match that, I'm still gonna complain.

I've told you about the forums before, and how they helped keep people
informed during patches. You refused to read them because you don't like
them. Fine, as I said at the time, your choice, but don't complain that
Sony doesn't update us, it does, just not in the place you're looking.

People who care about the game post the updates on the forums, people who
get paid peanuts to boot servers and update status pages don't bother.

[the wharf rat]

Post by Tony Evans
Click 'Forums' on the EQPlayers main page you love so much.

Interestingly enough, I just looked at "Forums" and what
did I find but a Forum for asking why the hell the network status
page doesn't have network status...

http://forums.station.sony.com/eq/posts/list.m?topic_id=116173

[Tony]

Post by the wharf rat

Post by Tony Evans
Click 'Forums' on the EQPlayers main page you love so much.

Interestingly enough, I just looked at "Forums" and what
did I find but a Forum for asking why the hell the network status
page doesn't have network status...
http://forums.station.sony.com/eq/posts/list.m?topic_id=116173

Yes, lots of people are upset about the lack of useful information on
EQPlayers. However, they post their concerns on the Sony site, where
Sony will see them. That's marginally more constructive than moaning
every single time Sony patches in a forum that Sony doesn't read.

There are three reasons I can see for moaning in here,

1. in the hope of warning other people
2. in the hope of getting it fixed
3. just to vent because it frustrates you

Well, #1 we all know it's broken, #2 no one who can fix it reads this
forum and #3 can be also resolved by checking somewhere that *does* get
updated.

So that leaves me with

#4 Likes to moan and wants to just cry about something for the hell of it.

[Faned]

Post by Tony Evans

Post by Lance Berg
Post a link to this forum you claim has the accurate poop, and I'll
bookmark it and try looking there.

Click 'Forums' on the EQPlayers main page you love so much.

EQPlayers? Obviously you have to have inside info to start from there.
Give me the links necessary to get there from www.everquest.com. =)

[Tony]

Post by Faned

Post by Tony Evans

Post by Lance Berg
Post a link to this forum you claim has the accurate poop, and I'll
bookmark it and try looking there.

Click 'Forums' on the EQPlayers main page you love so much.

EQPlayers? Obviously you have to have inside info to start from there.
Give me the links necessary to get there from www.everquest.com. =)

Click the word 'Forums'.

Yeh I appreciate it was complex.

[Pris Flynn]

Post by Tony

Post by Faned

Post by Tony Evans

Post by Lance Berg
Post a link to this forum you claim has the accurate poop, and I'll
bookmark it and try looking there.

Click 'Forums' on the EQPlayers main page you love so much.

EQPlayers? Obviously you have to have inside info to start from there.
Give me the links necessary to get there from www.everquest.com. =)

Click the word 'Forums'.
Yeh I appreciate it was complex.
--
Tony Evans
Saving trees and wasting electrons since 1993
blog -> http://perception-is-truth.blogspot.com/
[ anything below this line wasn't written by me ]

I do not see the "Forum"s link on the www.everquest.com link. However to
the far right I see half of the word Everquest, which does link to the
EQPlayers website. From there, Forums can be selected.

[Faned]

Post by Tony

Post by Faned

Post by Tony Evans

Post by Lance Berg
Post a link to this forum you claim has the accurate poop, and I'll
bookmark it and try looking there.

Click 'Forums' on the EQPlayers main page you love so much.

EQPlayers? Obviously you have to have inside info to start from there.
Give me the links necessary to get there from www.everquest.com. =)

Click the word 'Forums'.
Yeh I appreciate it was complex.

Why would I click on 'Forums' to see network status? The most non-technical
among us realizes that the forums aren't actually connected directly to the
game servers.

Again you are assuming inside info. Try to approach the situation as if you
didn't know where to find the network status.

[Tony]

Post by Faned
Again you are assuming inside info. Try to approach the situation as if you
didn't know where to find the network status.

I can't, it's impossible to know how I would react knowing nothing.

That's not the point I have at any stage been trying to make.

Lance did know better, because I told him a long time back, he ignored
that advice and continued to get frustrated about the lack of updates in
the one place he kept checking.

Sony's public EQ face is shockingly bad. It's communication is bad.
The way it manages the game is bad. I agree with all these things.

However, that doesn't preclude us from taking action ourselves to
alleviate how much frustration we suffer. I don't suffer any
frustration at the EQ Players page because I don't use it.

[Faned]

Post by Tony

Post by Faned
Again you are assuming inside info. Try to approach the situation as if you
didn't know where to find the network status.

I can't, it's impossible to know how I would react knowing nothing.

It's really pretty easy. Try to log in during a patch to create a new
character (since you won't get in, the "to create a new character" part is
irrelevant, but is included for "frame of mind"). When you fail, do the
logical thing and go to www.everquest.com. Start clicking and make note of
when and where you find a path that gets you to an accurate network status.
Hardly impossible.

Post by Tony
That's not the point I have at any stage been trying to make.
Lance did know better, because I told him a long time back, he ignored
that advice and continued to get frustrated about the lack of updates in
the one place he kept checking.

I went to my bank and asked what my balance was. They said to look in the
forums on their website.

I went to Walmart and asked how much a 24 pack of Mountain Dew was. They
said to look in the forums on their website.

I stopped at a used car lot and asked how much a 2002 Volkswagen Jetta they
had was. They told me to look in the forums on their website.

Being told where to look does not make not having the information in an
illogical place any less illogical.

Post by Tony
Sony's public EQ face is shockingly bad. It's communication is bad.
The way it manages the game is bad. I agree with all these things.
However, that doesn't preclude us from taking action ourselves to
alleviate how much frustration we suffer. I don't suffer any
frustration at the EQ Players page because I don't use it.

I can simply see Lance's point. I'm at work while EQ is patching, so it
almost never impacts me and I've never bothered to go any farther than the
EQ patcher to find out whether I can log in or not.

[Tony Evans]

Post by Faned
I went to my bank and asked what my balance was. They said to look in the
forums on their website.
I went to Walmart and asked how much a 24 pack of Mountain Dew was. They
said to look in the forums on their website.
I stopped at a used car lot and asked how much a 2002 Volkswagen Jetta they
had was. They told me to look in the forums on their website.
Being told where to look does not make not having the information in an
illogical place any less illogical.

Did you call your bank a second time and ask again how to find your
balance?

And a third.

And a fourth?

After each time you were told the forums were the place to go did you
continue going to the other place and telling the folk you see in the pub
that they still don't work?

After how many goes of that did they get bored and tell you to shut up?

[Faned]

Post by Tony Evans

Post by Faned
I went to my bank and asked what my balance was. They said to look in the
forums on their website.
I went to Walmart and asked how much a 24 pack of Mountain Dew was. They
said to look in the forums on their website.
I stopped at a used car lot and asked how much a 2002 Volkswagen Jetta they
had was. They told me to look in the forums on their website.
Being told where to look does not make not having the information in an
illogical place any less illogical.

Did you call your bank a second time and ask again how to find your
balance?

Yes, because that post wasn't updated after I made a deposit. They told me
to look at the new post.

Post by Tony Evans
And a third.

Yes, because after I paid my car payment I looked for a new post and there
wasn't one. They told me to check the original post because it had been
edited and updated.

Post by Tony Evans
And a fourth?

Yes, because now I have two or three different posts to check and I'm never
sure whether one of those will be updated or a completely new post will be
made with my current balance.

Post by Tony Evans
After each time you were told the forums were the place to go did you
continue going to the other place and telling the folk you see in the pub
that they still don't work?

The people in the pub don't care whether I can find out my account balance
or not. Doesn't stop me from complaining over a beer because my bank is
doing something stupid.

Post by Tony Evans
After how many goes of that did they get bored and tell you to shut up?

Didn't take many. But then I kicked the guy's ass, because he was the one
that told me to use that bank. =)

[Tony Evans]

Post by Faned
Didn't take many. But then I kicked the guy's ass, because he was the one
that told me to use that bank. =)

And smart asses should read more carefully.

It's not a POST on the forums. It's an ANNOUNCEMENT in the announcement
section in the top which is kept up-to-date.

[D.J.]

On Thu, 02 Aug 2007 20:36:55 +0100, Tony Evans
<postmaster@[127.0.0.1]> wrote:
]In alt.games.everquest, Faned <***@quellious.eq> wrote:
]
]>Didn't take many. But then I kicked the guy's ass, because he was the one
]>that told me to use that bank. =)
]
]And smart asses should read more carefully.
]
]It's not a POST on the forums. It's an ANNOUNCEMENT in the announcement
]section in the top which is kept up-to-date.

He isn't listening. He doesn't appear to want to learn how to find
out if the servers are up or not.

JimP.

[Eric D. Braden]

Post by Faned

Post by Tony

Post by Faned
Again you are assuming inside info. Try to approach the situation as if you
didn't know where to find the network status.

I can't, it's impossible to know how I would react knowing nothing.

It's really pretty easy. Try to log in during a patch to create a new
character (since you won't get in, the "to create a new character" part is
irrelevant, but is included for "frame of mind"). When you fail, do the
logical thing and go towww.everquest.com. Start clicking and make note of
when and where you find a path that gets you to an accurate network status.
Hardly impossible.

Post by Tony
That's not the point I have at any stage been trying to make.
Lance did know better, because I told him a long time back, he ignored
that advice and continued to get frustrated about the lack of updates in
the one place he kept checking.

I went to my bank and asked what my balance was. They said to look in the
forums on their website.

Am I the only one who considers official forums a much better source
of info than "official sites"? I've never even started to look for
the realm status page. Forums are almost always the best place to
look for any current info, because even if there is no official word,
there will at least be people talking about it.

This isn't really useful for the conversation at hand, but it does
seem like some people are asserting that they didn't even know the
forums existed. Am I odd in that forums are one of the first things I
look for in a new online game?

[Tony Evans]

Post by Eric D. Braden
This isn't really useful for the conversation at hand, but it does
seem like some people are asserting that they didn't even know the
forums existed. Am I odd in that forums are one of the first things I
look for in a new online game?

Well at the very least I don't continuously follow the same route to a page
I know doesn't work.

I always check the forums for games I'm playing, because they invariably
have knowledgable posters.

[D.J.]

On Thu, 02 Aug 2007 08:58:51 -0500, Faned <***@quellious.eq>
wrote:
]<tony@[127.0.0.1]> wrote:
]> Faned wrote:
]> > <postmaster@[127.0.0.1]> wrote:
]> >> In alt.games.everquest, Lance Berg <***@NOSPAM.dejazzd.com> wrote:
]> >>
]> >>> Post a link to this forum you claim has the accurate poop, and I'll
]> >>> bookmark it and try looking there.
]> >> Click 'Forums' on the EQPlayers main page you love so much.
]> >
]> > EQPlayers? Obviously you have to have inside info to start from there.
]> > Give me the links necessary to get there from www.everquest.com. =)
]>
]> Click the word 'Forums'.
]>
]> Yeh I appreciate it was complex.
]
]Why would I click on 'Forums' to see network status? The most non-technical
]among us realizes that the forums aren't actually connected directly to the
]game servers.

Actually, we are presuming you can follow simple directions.

Stop sticking your fingers in your ears and going 'lalalalalala !'.

]Again you are assuming inside info. Try to approach the situation as if you
]didn't know where to find the network status.

You have been told where to find it. No inside info needed.

JimP.

[Faned]

Post by D.J.
]> >>
]> >>> Post a link to this forum you claim has the accurate poop, and I'll
]> >>> bookmark it and try looking there.
]> >> Click 'Forums' on the EQPlayers main page you love so much.
]> >
]> > EQPlayers? Obviously you have to have inside info to start from there.
]> > Give me the links necessary to get there from www.everquest.com. =)
]>
]> Click the word 'Forums'.
]>
]> Yeh I appreciate it was complex.
]
]Why would I click on 'Forums' to see network status? The most non-technical
]among us realizes that the forums aren't actually connected directly to the
]game servers.
Actually, we are presuming you can follow simple directions.

You told me to click on 'Forums' to see network status. When I click on
'Forums', I'm going to see forums. Your instructions make no sense. =P

Post by D.J.
Stop sticking your fingers in your ears and going 'lalalalalala !'.

"Lalalalala! Don't use the network status page to check network status!
Look at some ill defined spot in the forums! Lalalalala!"

Post by D.J.
]Again you are assuming inside info. Try to approach the situation as if you
]didn't know where to find the network status.
You have been told where to find it. No inside info needed.

I haven't tried yet. I might tonight just to find out how convoluted the
process of finding out the true network status is.

But anyway, being told pretty much constitutes inside info...

[Richard]

Post by Tony Evans

Post by the wharf rat
Because the network status page includes directions in 24 point
type instructing you to stop expecting the network status page to
contain network status and go find The Forums? (Which Forum did you
say it was by the way? "Test", or something equally intuitive, wasn't
it?)

No, it's THE Everquest Forums.
The front page of them.
Start with EQPlayers.com (if you must), click Forums, read the text in
the box which at the moment says "The EverQuest Servers are up!~
Zatozia" but early gave information on the servers being down.
Not exactly hard to find.

Post by the wharf rat
Yes, I'm familiar with this attitude. "It was difficult to
design so it should be difficult to use". How quaint to cling to such
thinking in this modern era. Sony and its apologists: coelocanths
sculling across the slimy bottom of the sea of service...

Huh? I agreed that the eqplayers stuff sucks, I'm saying there IS an
option that works, which is EASY to use and find, and that by NOT
using it and then complaining about the one everyone knows doesn't
work Lance is being intentionally obtuse.

Reading this here and now is the first time I heard about them posting it
in the forums. Now that I know, I will. It is stupid, but then, it is
SOE.

[D.J.]

On Tue, 31 Jul 2007 23:07:10 +0000 (UTC), ***@panix.com (the wharf
rat) wrote:
]In article <46afb201$0$648$***@news.gradwell.net>,
]Tony Evans <${tony}$@darkstorm.co.uk> wrote:
]>
]>They DO post the information in an EASY to read place on the FRONT page of
]>the Forums *usually*. It's very useful, it's very easy to check.
]>
]
] And you know this how?

look at this page every day. No need to log in to anything.
If the servers are down for patch, it will say so.
http://forums.station.sony.com/eq/forums/list.m

No need for you to feel picked on. SOE doesn't care if you exist or
not.

JimP.

[Lief]

Post by D.J.
]>
]>They DO post the information in an EASY to read place on the FRONT page of
]>the Forums *usually*. It's very useful, it's very easy to check.
]>
]
] And you know this how?
look at this page every day. No need to log in to anything.
If the servers are down for patch, it will say so.
http://forums.station.sony.com/eq/forums/list.m
No need for you to feel picked on. SOE doesn't care if you exist or
not.

Who cares about SoE and what they do? All irrelevant, I played since 2002
on and off and dont give a crap what they do really (not that I wont moan a
bit about ogm my class needs this, etc, but it's not like it's gonna stop me
playing), and seen people who have played even longer, do nothing but
complain about SoE and how they hate us all etc and how the game is so bad
yadayadayada. But they continue to play it, and, somehow manage to have
fun.............

There is no win for SoE, patch, people complain, dont patch, people
complain, 6 hours is not exactly a long time, to not be able to play A GAME,
for christ sake.

Even if you dont know, you try and log in 'oh nos I cant log in' and it
tells you it's patching, finding out the current patch info is not rocket
science.

If you aint having fun, fuck off to World of Holdyourhand.

Hmm, 2am pointless rants win!!

(not pointed at you btw D.J, just a general rant).

[the wharf rat]

Post by Lief
There is no win for SoE, patch, people complain, dont patch, people
complain, 6 hours is not exactly a long time, to not be able to play A GAME,
for christ sake.

We're not complaining about the downtime. We're complaining
about the inconsiderate and unnecessary worse-than-lack of the kind of basic
customer service even Ma&Pa ISP L.L.C. can provide. (Well, at least I am)

I'm also complaining about the kind of slapdash quality control
that rolls out a massive patch to critical functionality without even testing
it. I know it's just a game. It's the principal of the thing, really. To
paraphrase Horton the Elephant, "Bullshit is Bullshit, no matter how small".

It's also hard not to at least roll your eyes over nonsense like
network status pages that don't have network status, isn't it? Or are
you immune to irony?

[Lief]

Post by the wharf rat

Post by Lief
There is no win for SoE, patch, people complain, dont patch, people
complain, 6 hours is not exactly a long time, to not be able to play A GAME,
for christ sake.

We're not complaining about the downtime. We're complaining
about the inconsiderate and unnecessary worse-than-lack of the kind of basic
customer service even Ma&Pa ISP L.L.C. can provide. (Well, at least I am)

I've seen worse CS :p

Post by the wharf rat
I'm also complaining about the kind of slapdash quality control
that rolls out a massive patch to critical functionality without even testing
it. I know it's just a game. It's the principal of the thing, really.

To

Post by the wharf rat
paraphrase Horton the Elephant, "Bullshit is Bullshit, no matter how small".

It is tested, of course they cannot spend a year on testing, so it's given
to test server, tried on there. I think considering the antiquity of the
code, and the bulk of it, in generally they do a good job.

Major problems are in general, fixed quickly, which is great for an old game
like this.

I'd love for every patch to go off without a hitch, but that would require a
perfect world ;p

Post by the wharf rat
It's also hard not to at least roll your eyes over nonsense like
network status pages that don't have network status, isn't it? Or are
you immune to irony?

It wasnt hard to find out the real patch time, that network status page has
always been useless...all it needs is to have one link at the top saying
'login server up / down'

[the wharf rat]

Post by Lief
It is tested, of course they cannot spend a year on testing, so it's given
to test server, tried on there. I think considering the antiquity of the
code, and the bulk of it, in generally they do a good job.

Pfffft :-)

They changed the AA code massively, rolled it out, and then realized
that AAs no longer worked.

The problem is that they DON'T test anything. They put it on a test
server and hope one of the customers notices a problem and can explain what's
wrong in a way that they can understand and care about. That's a broken
approach, but it fails completely when you're dealing with functionality
that's deliberately hidden from the players (such as proper AA behavior).

I dunno, I guess I feel that if more people objected to this it
might change but it's like Windows blue screens: people are so used to
them that they think your operating system is *supposed* to be broken...

Post by Lief
I'd love for every patch to go off without a hitch, but that would require a
perfect world ;p

I'd love for ONE patch to go off without a hitch.

[Tony]

Post by the wharf rat
I dunno, I guess I feel that if more people objected to this it
might change ...

Go to the Sony boards, post constructive feedback. Write to smedley,
provide constructive feedback. Start the process.

[the wharf rat]

Post by Tony
Go to the Sony boards, post constructive feedback. Write to smedley,
provide constructive feedback. Start the process.

I would if I thought they cared. They don't or in 9 years
they'd have instituted industry standard QA and operational control
procedures.

But hey, we all know I'm just whining so why bother?

[Tony Evans]

Post by the wharf rat
But hey, we all know I'm just whining so why bother?

Agreed.

[Richard Carpenter]

Post by the wharf rat

Post by Tony
Go to the Sony boards, post constructive feedback. Write to smedley,
provide constructive feedback. Start the process.

I would if I thought they cared. They don't or in 9 years
they'd have instituted industry standard QA and operational control
procedures.

I have little doubt that it has always been about prioritization of
resources. Their "QA and operational control procedures" exist. How close
it comes to meeting "industry standards" appears to be quite subjective.

[Richard Carpenter]

Post by the wharf rat

Post by Lief
There is no win for SoE, patch, people complain, dont patch, people
complain, 6 hours is not exactly a long time, to not be able to play A
GAME, for christ sake.

We're not complaining about the downtime. We're complaining
about the inconsiderate and unnecessary worse-than-lack of the kind of
basic customer service even Ma&Pa ISP L.L.C. can provide. (Well, at
least I am)
I'm also complaining about the kind of slapdash quality control
that rolls out a massive patch to critical functionality without even
testing it. I know it's just a game. It's the principal of the
thing, really. To paraphrase Horton the Elephant, "Bullshit is
Bullshit, no matter how small".

Oh come on. Do you really believe that they roll out major fixes without
any testing? That's absurd. Of course they test it. People just fail to
realize that the testing performed by the few people at the EQ dev
offices and then by the not-many-more who play on Test doesn't begin to
compare to the ringer such changes are put through once they are
implemented on Live servers. For every change that is so lamented in the
forums for it's perceived haphazard testing and rollout, I would
honestly bet that there are a hundred that go in without any trouble at
all. We just don't tend to hear much about those, though, do we? ;)

Post by the wharf rat
It's also hard not to at least roll your eyes over nonsense like
network status pages that don't have network status, isn't it? Or are
you immune to irony?

Oh, plenty of eyes have rolled about this over the years. The bottom
line is that it's been this way for a long time now (not that that
should mean it's ok - just that it shouldn't be getting anyone worked up
at this point). That combined with the fact that there is another more
reliable resource makes me inclined to agree with the "it's no big
deal" crowd. Under the circumstances, it really is no big deal. I think
the only reason people seem to be balking at the notion that it should
be fixed is the fact that it seems to be brought up time and time again
by the same people (no offense, Lance), leading others to believe that
they're just getting their undies in a bunch unnecessarily.

I'm certainly not one to ignore the shortcomings of SoE (EQ is just
about the only Sony product I'll consume, with the exception of movies -
man, the Sony name seems to be on half the films I see, anymore), but I
do try to maintain some perspective.

[Faned]

Post by Richard Carpenter

Post by the wharf rat

Post by Lief
There is no win for SoE, patch, people complain, dont patch, people
complain, 6 hours is not exactly a long time, to not be able to play A
GAME, for christ sake.

We're not complaining about the downtime. We're complaining
about the inconsiderate and unnecessary worse-than-lack of the kind of
basic customer service even Ma&Pa ISP L.L.C. can provide. (Well, at
least I am)
I'm also complaining about the kind of slapdash quality control
that rolls out a massive patch to critical functionality without even
testing it. I know it's just a game. It's the principal of the
thing, really. To paraphrase Horton the Elephant, "Bullshit is
Bullshit, no matter how small".

Oh come on. Do you really believe that they roll out major fixes without
any testing? That's absurd. Of course they test it. People just fail to
realize that the testing performed by the few people at the EQ dev
offices and then by the not-many-more who play on Test doesn't begin to
compare to the ringer such changes are put through once they are
implemented on Live servers.

Demonstrably false, throughout the history of the game. But let's just take
this past month as an example. The AA's were combined. It didn't work.
The "testing" needed to figure that out was simply logging in and trying one
of the changed AAs. Because it is blindingly obvious that that wasn't done,
it is safe to say that there was absolutely zero testing.

I don't really complain about it anymore. The only time I'll enter such a
conversation is when someone tosses out a pie-in-the-sky "Do you really
believe that they roll out major fixes without any testing? That's absurd.
Of course they test it." When that happens, I have to point out how it
conflicts with reality. =)

[Tony]

Post by Faned
Demonstrably false, throughout the history of the game. But let's just take
this past month as an example. The AA's were combined. It didn't work.
The "testing" needed to figure that out was simply logging in and trying one
of the changed AAs. Because it is blindingly obvious that that wasn't done,
it is safe to say that there was absolutely zero testing.

Ridiculous assertion. There could have been 99.9% testing complete and
they didn't test the one AA you checked. You've demonstrated nothing.

Look, I'm not a sony apologist, I think their current Formal
communication is terrible, their latest producer's newsletter a complete
waste of space, their marketing is crap but at least let's be honest.
They test, but they don't test well enough, maybe due to resources,
maybe due to skill, maybe due to other reasons, but to deny any testing
at all takes place is just to be on the other extreme of fan boy.

[the wharf rat]

Post by Tony
Ridiculous assertion. There could have been 99.9% testing complete and
they didn't test the one AA you checked. You've demonstrated nothing.

It wasn't one AA. It was every one that they combined using
their new scheme.

Post by Tony
Look, I'm not a sony apologist

Sure you are.

[Tony Evans]

Post by the wharf rat
It wasn't one AA. It was every one that they combined using
their new scheme.

No, it wasn't 'every one'. And choosing the extreme just burns your
argument. It was many of them, some of them, a number of them, but it
wasn't every single line of combined AA.

Post by the wharf rat

Post by Tony
Look, I'm not a sony apologist

Sure you are.

No I'm not. Itemisation sucks, the current item Dev needs to be replaced.
I've PM'd the producer about that on the boards, I've written posts on the
boards explaining why TBS Warrior Armour is badly itemised and providing
constructive examples on how to fix it.

Your black and white picture is inaccurate.

[Faned]

Post by Tony Evans

Post by the wharf rat
It wasn't one AA. It was every one that they combined using
their new scheme.

No, it wasn't 'every one'. And choosing the extreme just burns your
argument. It was many of them, some of them, a number of them, but it
wasn't every single line of combined AA.

Are you sure about that? From the sheer number mentioned in the patch
message (as well as the others that were fixed and weren't mentioned), I
gathered that the problem was pretty much across the board. If it wasn't
all of them, the ones that weren't broken were the exceptions.

Post by Tony Evans

Post by the wharf rat

Post by Tony
Look, I'm not a sony apologist

Sure you are.

No I'm not. Itemisation sucks, the current item Dev needs to be replaced.
I've PM'd the producer about that on the boards, I've written posts on the
boards explaining why TBS Warrior Armour is badly itemised and providing
constructive examples on how to fix it.
Your black and white picture is inaccurate.

Black = Sony doesn't test
White = Sony tests

[Tony]

Post by Faned
Are you sure about that?

Yes, my warrior is at max AA, I know how many of those AA continued to
work. My wizard has 500 AA, I know how many of those continued to work.

Post by Faned

Post by Tony Evans
Your black and white picture is inaccurate.

Black = Sony doesn't test
White = Sony tests

Black - Sony tests nothing
White - Sony tests everything
Reality - Sony tests some stuff

[Faned]

Post by Tony

Post by Faned
Are you sure about that?

Yes, my warrior is at max AA, I know how many of those AA continued to
work. My wizard has 500 AA, I know how many of those continued to work.

Which of the newly combined, additive AAs continued to work? Serious
question, as I'm not aware of any that did. I know in my personal
experience the critical hit/heal AAs and attack enhancements (extra double
attack, etc.) were all broken.

Post by Tony

Post by Faned

Post by Tony Evans
Your black and white picture is inaccurate.

Black = Sony doesn't test
White = Sony tests

Black - Sony tests nothing
White - Sony tests everything
Reality - Sony tests some stuff

I agree, they test whether a patch will crash the servers or the clients. I
see some evidence of that (though even that isn't perfect). They don't test
much more than that (as is plainly obvious every time something shows up on
live servers that never landed on the test server first).

[Faned]

Post by Tony

Post by Faned
Demonstrably false, throughout the history of the game. But let's just take
this past month as an example. The AA's were combined. It didn't work.
The "testing" needed to figure that out was simply logging in and trying one
of the changed AAs. Because it is blindingly obvious that that wasn't done,
it is safe to say that there was absolutely zero testing.

Ridiculous assertion. There could have been 99.9% testing complete and
they didn't test the one AA you checked. You've demonstrated nothing.

You fail to comprehend how badly they don't test. It wasn't "the one AA"
that I checked that didn't work. It was most AAs (if not all) that were
combined that didn't work. As in, they only had to test one, basically any
one, and they should have been able to discern that there was a problem. I
personally had at least three broken AAs that were visibly and noticeably
broken, and I *didn't* "test".

Post by Tony
Look, I'm not a sony apologist, I think their current Formal
communication is terrible, their latest producer's newsletter a complete
waste of space, their marketing is crap but at least let's be honest.
They test, but they don't test well enough, maybe due to resources,
maybe due to skill, maybe due to other reasons, but to deny any testing
at all takes place is just to be on the other extreme of fan boy.

You fail to comprehend something else. I'm a fan boy. I've played EQ since
it started and will probably be the last to log off when the servers go
dark. I like the game. That doesn't blind me to the facts.

There are a seemingly endless string of similar stories peppered throughout
the history of the game. I think the one that convinced me that it was
worse than bad testing, but was in fact no testing, was when they broke bard
songs *entirely* one patch when the patch message specified some small
change to bards. It was completely obvious that nobody had done even the
slightest bit of testing. It took about 30 seconds after the servers came
up for the posts to start showing up on the boards complaining about the
fact that bards couldn't play any songs at all.

You can pretend all you want that they test. I know from a preponderance of
evidence that they do not.

[Richard Carpenter]

Post by Faned

Post by Richard Carpenter

Post by the wharf rat

Post by Lief
There is no win for SoE, patch, people complain, dont patch, people
complain, 6 hours is not exactly a long time, to not be able to
play A GAME, for christ sake.

We're not complaining about the downtime. We're complaining
about the inconsiderate and unnecessary worse-than-lack of the kind
of basic customer service even Ma&Pa ISP L.L.C. can provide.
(Well, at least I am)
I'm also complaining about the kind of slapdash quality control
that rolls out a massive patch to critical functionality without
even testing it. I know it's just a game. It's the principal of
the thing, really. To paraphrase Horton the Elephant, "Bullshit is
Bullshit, no matter how small".

Oh come on. Do you really believe that they roll out major fixes
without any testing? That's absurd. Of course they test it. People
just fail to realize that the testing performed by the few people at
the EQ dev offices and then by the not-many-more who play on Test
doesn't begin to compare to the ringer such changes are put through
once they are implemented on Live servers.

Demonstrably false, throughout the history of the game. But let's
just take this past month as an example. The AA's were combined. It
didn't work. The "testing" needed to figure that out was simply
logging in and trying one of the changed AAs. Because it is
blindingly obvious that that wasn't done, it is safe to say that there
was absolutely zero testing.
I don't really complain about it anymore. The only time I'll enter
such a conversation is when someone tosses out a pie-in-the-sky "Do
you really believe that they roll out major fixes without any testing?
That's absurd. Of course they test it." When that happens, I have to
point out how it conflicts with reality. =)

If such practices were common, you would see client crashes resulting,
rather than mere malfunctioning features. Especially with such a
significant change as was made to the AA system, do you really think
that none of the developers even ran the program and triggered the AA's
to make sure nothing was overlooked that might immediately crash the
game?

I would agree that *more* testing was warranted, but to make the claim
that these changes are rolled out with *no* testing is just pure
sensationalism and blatantly false.

Out of curiousity, weren't these changes rolled out on the test server
prior to going live?

[the wharf rat]

Post by Richard Carpenter
If such practices were common, you would see client crashes resulting,
rather than mere malfunctioning features.

That's a non sequitur.

Post by Richard Carpenter
I would agree that *more* testing was warranted, but to make the claim
that these changes are rolled out with *no* testing is just pure
sensationalism and blatantly false.

Ok, if they tested them why didn't they notice it didn't work?
This isn't a complicated and difficult to reproduce interaction between
unknown user input and a rare game engine state. It's just applying the
values from a static data structure to an arithmetic calculation.

This is the kind of mistake that would get caught in code review.
If they did code reviews, that is.

[Richard Carpenter]

Post by the wharf rat

Post by Richard Carpenter
If such practices were common, you would see client crashes resulting,
rather than mere malfunctioning features.

That's a non sequitur.

Um, no, it's not. That most certainly *is* the sort of thing that is
very likely to happen when code is manipulated without the least bit of
testing. I know of which I speak. If nothing else, programmers will run
the code as a safeguard against simple syntax errors.

Post by the wharf rat

Post by Richard Carpenter
I would agree that *more* testing was warranted, but to make the claim
that these changes are rolled out with *no* testing is just pure
sensationalism and blatantly false.

Ok, if they tested them why didn't they notice it didn't work?

Who knows? Certainly neither you nor I. What is much more certain,
however, is that they don't just throw in redesigns of such magnitude
without *any* testing. Not a chance. My guess is, as I suggested, the
testing they *did* was merely not as thorough as it should have been.
Perhaps the person (it's quite possible there was only one, again, who
knows) responsible for testing this new mechanism, didn't do much beyond
triggering the AA's and making sure the text popped up in the chat
window and the game didn't crash as a result. Maybe they just assumed in
that case that all was working.

Post by the wharf rat
This isn't a complicated and difficult to reproduce interaction
between unknown user input and a rare game engine state. It's just
applying the values from a static data structure to an arithmetic
calculation.

You're applying pie in the sky requirements to a shoestring budget,
9-year-old MMORPG. For crying out loud, the resources necessary to come
up with the sort of testing facilities upon which you insist should be
in place for *all* program functions would quite likely require more
manpower than they have devoted to the entire game at this time. I
really think you need to temper your expectations a bit.

Post by the wharf rat
This is the kind of mistake that would get caught in code review.
If they did code reviews, that is.

Again, we don't know whether they do or not. Debating that point from
either side would be...well...pointless. It certainly shouldn't be used
to bolster an argument.

Bottom line: anyone who says they implement game redesigns without any
testing is either loopy or just being a malcontent.

[the wharf rat]

Post by Richard Carpenter

Post by the wharf rat

Post by Richard Carpenter
If such practices were common, you would see client crashes resulting,
rather than mere malfunctioning features.

That's a non sequitur.

Um, no, it's not. That most certainly *is* the sort of thing that is
very likely to happen when code is manipulated without the least bit of
testing.

Joining two clauses with a comma does not make them logically
connected. Lack of testing means that defects will appear in production.
It doesn't predict what those defects will be or where they will appear.

To do that you'd have to TEST.

Post by Richard Carpenter
I know of which I speak. If nothing else, programmers will run
the code as a safeguard against simple syntax errors.

Yup. Exactly the attitude that's getting Sony in trouble.
"It compiles. Ship it." They're not the first to take that attitude
and they won't be the last, but it's wrong on many levels.


BTW, we *do* see client crashes. I can crash my client simply
by leaving my trader up for a long enough time then trying to switch
characters. I can also crash it by switching back to the character
selection screen "enough". It's reproducible enough so that I always
log back to the desktop before doing anything important. That's not the
kind of bug I'm talking about. I'm talking about rolling out production
releases that are obviously broken in trivially TESTABLE ways. All new AAs
don't work. No songs work. Spell gems don't refresh unless you zone.

Post by Richard Carpenter

Post by the wharf rat
Ok, if they tested them why didn't they notice it didn't work?

Who knows? Certainly neither you nor I.

I do. They didn't notice because they didn't look. Because
if they HAD looked there's NO WAY they could have missed it.

Post by Richard Carpenter
What is much more certain,
however, is that they don't just throw in redesigns of such magnitude
without *any* testing. Not a chance.

Oh, think so?

Post by Richard Carpenter
Maybe they just assumed in that case that all was working.

Yes, exactly. They don't bother testing because it's just
a game, testing sucks, and anyway it's time for foosball. It's still
1991 at SOE.

Post by Richard Carpenter
You're applying pie in the sky requirements to a shoestring budget,

Bullshit. Proper QA saves tons of money. It costs 100 times
as much to fix a bug that's been rolled out to production as it does
to fix it in system test, and it costs 10 times as much to fix it in
system test as it does to fix it in unit test. If someone at Sony
demanded they stop and get this right you'd all get your 25 new zones hardly
anyone will ever see each and every year and the rest of us might not have
to whine about the way hill giants warp when you try and kite them.

Post by Richard Carpenter
Bottom line: anyone who says they implement game redesigns without any
testing is either loopy or just being a malcontent.

Got some evidence of that statement, you loopy malcontent who
doesn't seem to understand the first thing about proper software engineering?
(You're right! Gratuitous insults are fun! And here I was wasting my time
trying to keep this on an adult level...)

[Schadenfreude]

Post by the wharf rat
BTW, we *do* see client crashes. I can crash my client simply
by leaving my trader up for a long enough time then trying to switch
characters. I can also crash it by switching back to the character
selection screen "enough". It's reproducible enough so that I always
log back to the desktop before doing anything important. That's not the
kind of bug I'm talking about

I leave my trader up for ages, sometimes two traders and sometimes one
trader and one active character on the same PC and switch between
multiple characters on multiple accounts all the time and very rarely
have *any* sort of issues despite multiple EQ instances on one PC
being completely 100% unsupported.

You need to differentiate between client bugs and crap that happens on
your PC. The West bug for example is an obscure but verifiable client
problem with certain video cards at certain times but does happen.

[the wharf rat]

Post by Schadenfreude
You need to differentiate between client bugs and crap that happens on
your PC.

I'm reasonably sure it's a memory leak in the client, because of
the behavior. The PC is a well maintained vanilla Intel XP box with 100%
updated patches and drivers, and no special hardware. It also does nothing
but run EQ. I do all my real work on Linux :-)

Sure, it might be local driver bug. That's why stuff like this
doesn't upset me (on Windows :-). It's almost not worth the effort to find
and fix. Glaringly obvious critical bugs in a production software release
are different kettle of cobalt cod.

[D.J.]

On Sat, 4 Aug 2007 15:04:53 +0000 (UTC), ***@panix.com (the wharf
rat) wrote:
]In article <***@4ax.com>,
]Schadenfreude <***@hotmail.com> wrote:
]>
]>You need to differentiate between client bugs and crap that happens on
]>your PC.
]
] I'm reasonably sure it's a memory leak in the client, because of

From I have read in various places, it is a memory leak.

It has gotten worse. I used to be able to play a toon on one server,
switch to another server, start a new toon on another server. Then
got back to the first server and play there. But not recently.

JimP.

[Richard Carpenter]

Post by the wharf rat

Post by Schadenfreude
You need to differentiate between client bugs and crap that happens on
your PC.

I'm reasonably sure it's a memory leak in the client, because of
the behavior.

Then please explain, oh oracle of all things coded, why it doesn't affect
Shadenfreude's or my PC in that manner? A memory leak would be universal to
all clients.

[D.J.]

On Sat, 04 Aug 2007 13:05:52 -0500, Richard Carpenter
<***@hotspamsuxmail.com> wrote:
]***@panix.com (the wharf rat) wrote in
]news:f924il$4a2$***@reader2.panix.com:
]
]> In article <***@4ax.com>,
]> Schadenfreude <***@hotmail.com> wrote:
]>>
]>>You need to differentiate between client bugs and crap that happens on
]>>your PC.
]>
]> I'm reasonably sure it's a memory leak in the client, because of
]> the behavior.
]
]Then please explain, oh oracle of all things coded, why it doesn't affect
]Shadenfreude's or my PC in that manner? A memory leak would be universal to
]all clients.

I remember a mention of a memory leak on the official forums.

But frankly, SOE's 'quality assurance' is anything but.

JimP.

[the wharf rat]

Post by Richard Carpenter
Then please explain, oh oracle of all things coded, why it doesn't affect
Shadenfreude's or my PC in that manner? A memory leak would be universal to
all clients.

Because, grasshopper, it might require a specific code path to
trigger either the leak (or to trigger it enough times to cause memory
problems) or to trigger the bug that causes the program to crash when a
memory request goes unsatisfied (which is actually a separate bug). The
effects of the leak could depend on the memory configuration of the host
system, the actual memory map (what's loaded where and when), or even the
precise level of OS patching.

Interestingly enough, there are commercial packages that will
instrument your code to find memory leaks (purify is a famous one, but
even the basic freeware gnu compiler comes with a trick malloc that does
useful things). So this might even be easy to fix.

[Richard Carpenter]

Post by the wharf rat

Post by Richard Carpenter
Then please explain, oh oracle of all things coded, why it doesn't
affect Shadenfreude's or my PC in that manner? A memory leak would be
universal to all clients.

Because, grasshopper, it might require a specific code path to
trigger either the leak (or to trigger it enough times to cause memory
problems) or to trigger the bug that causes the program to crash when
a memory request goes unsatisfied (which is actually a separate bug).
The effects of the leak could depend on the memory configuration of
the host system, the actual memory map (what's loaded where and when),
or even the precise level of OS patching.

What you've just described is a small possibility that would only be
manifested on a very specific PC configuration - hardly any sort of example
of the magnitude of dereliction you claim on SoE's part.

Admit it. Memory leaks, while being *promoted* by hardware is surely
possible, are 99% of the time the result of the program, itself, reserving
memory and not releasing it when finished with it, or releasing memory it
still needs and expects to be available. Therefore, explain to me, again,
why the number of crashing clients is inconsistent with that realistic
estimate...grasshoppa.

Go ahead and wax superior about all the years of experience you have under
your belt, but just understand that you're not the only programmer here.

By the way, the fact that you continue to *pay* to play a game you so
zealously criticize says far more than any of your self-indulgent postings.

[the wharf rat]

Post by Richard Carpenter
What you've just described is a small possibility that would only be
manifested on a very specific PC configuration - hardly any sort of example
of the magnitude of dereliction you claim on SoE's part.

How many times did I say "this isn't the kind of bug we're
looking for. You can pass" ? My Jedi abilities are not what they
used to be, I guess...

Post by Richard Carpenter
Admit it. Memory leaks, while being *promoted* by hardware is surely
possible, are 99% of the time the result of the program, itself, reserving
memory and not releasing it when finished with it, or releasing memory it
still needs and expects to be available. Therefore, explain to me, again,
why the number of crashing clients is inconsistent with that realistic
estimate...grasshoppa.

I can't parse this. What are you asking for?

Post by Richard Carpenter
Go ahead and wax superior about all the years of experience you have under
your belt, but just understand that you're not the only programmer here.

Well, I've managed to avoid doing so so far :-) Anyway, I'm
not talking about programming. I'm talking about quality control/quality
assurance. It's an entirely different discipline.

[Richard Carpenter]

Post by the wharf rat

Post by Richard Carpenter
What you've just described is a small possibility that would only be
manifested on a very specific PC configuration - hardly any sort of
example of the magnitude of dereliction you claim on SoE's part.

How many times did I say "this isn't the kind of bug we're
looking for. You can pass" ? My Jedi abilities are not what they
used to be, I guess...

Then why on earth did you feel the need to use it as an example?? Are you
just *trying* to confuse the issue?

Post by the wharf rat

Post by Richard Carpenter
Admit it. Memory leaks, while being *promoted* by hardware is surely
possible, are 99% of the time the result of the program, itself,
reserving memory and not releasing it when finished with it, or
releasing memory it still needs and expects to be available.
Therefore, explain to me, again, why the number of crashing clients is
inconsistent with that realistic estimate...grasshoppa.

I can't parse this. What are you asking for?

Try harder. I don't have time to explain plain english to you.

Post by the wharf rat

Post by Richard Carpenter
Go ahead and wax superior about all the years of experience you have
under your belt, but just understand that you're not the only
programmer here.

Well, I've managed to avoid doing so so far :-) Anyway, I'm
not talking about programming. I'm talking about quality
control/quality assurance. It's an entirely different discipline.

Could the hairs be split any finer?

[the wharf rat]

Post by Richard Carpenter
Then why on earth did you feel the need to use it as an example?? Are you
just *trying* to confuse the issue?

I used it as an example of the kind of bug that I wasn't
complaining about, remember? Geez.

Post by Richard Carpenter
Try harder. I don't have time to explain plain english to you.

Lol. That's not plain English. This is:

I don't think you have any idea what you're talking about.

[Richard Carpenter]

Post by the wharf rat

Post by Richard Carpenter
Then why on earth did you feel the need to use it as an example?? Are
you just *trying* to confuse the issue?

I used it as an example of the kind of bug that I wasn't
complaining about, remember? Geez.

I stated that the sort of negligence you were claiming would have often
resulted in client crashes and that we have not seen everyone crashing
after a patch (ever? dunno). You thought it necessary to also debate
whether or not we're seeing client crashes, obviously in the interest of
discounting that particular line of reasoning on my part.

Round and round we go. When you'll stop arguing for the sake of it, nobody
knows.

Post by the wharf rat

Post by Richard Carpenter
Try harder. I don't have time to explain plain english to you.

I don't think you have any idea what you're talking about.

Whatever. In your arrogance, you're not interested in hearing anything
inconsistent with your own views. You've already decided that you're right
beyond a shadow of a doubt, even though you have nothing but circumstantial
and unprivileged observation on which to base your claims. There's no hope
for discussion under these circumstances.

I'll just repeat it for the record: SoE does not roll out significant game
updates (such as the AA redesign) without *any* testing at all. To claim
bugs as proof that they do is a non sequitur. It's ridiculous.

[the wharf rat]

Post by Richard Carpenter
I stated that the sort of negligence you were claiming would have often
resulted in client crashes and that we have not seen everyone crashing
after a patch (ever? dunno). You thought it necessary to also debate
whether or not we're seeing client crashes, obviously in the interest of
discounting that particular line of reasoning on my part.

Well, no...

Look, I'll try to sum up my whining:

There's several classes of bugs you typically find in software.
One large class of those is the class that includes trivial programmer
errors that can be easily and inexpensively detected and corrected at
an early stage in the product life cycle by use of simple tools and
processes that have been standard best practice since, oh, 1995. I'm
offended, both professionally and as a customer, to see that SOE clearly
can't be bothered to implement simple best practice. That they don't
is abundantly evident by the classes of bugs they let out into production:
incorrect static initializations, incorrect arithmetic, bugs that were
fixed and then return over and over again... The truly sad thing is
that by implementing what everyone else has done for decades they'd spend
much less money on operations and could have that much more for development,
bug fixes, or their stockholders. And the customers would probably notice
the improvement, too.

Memory leaks are a different class of bug. There's not a lot of
point to chasing seldom occurring and hard to reproduce problems when
you can't even count on your data structures being initialized properly.
SOE rightly doesn't care about these. The, well, bug, is that they don't
care about anything else, either.

They don't test. "Testing" does not mean having your 7 year old
cousin run his twinked froglok wizard around PoK for 20 minutes before
you roll out the patch. Great Ghu, you can take a graduate degree in
Quality Control, did you know that? It's a regular *science*. And it's
*not* programming. It's more like applied statistics...

And speaking of bugs, when are they going to fix the bald druid
bug? When did that appear, 2002? Geez.

[Don Woods]

Post by the wharf rat
... The truly sad thing is
that by implementing what everyone else has done for decades they'd spend
much less money on operations and could have that much more for development,
bug fixes, or their stockholders. And the customers would probably notice
the improvement, too.

I have no doubt that if they had designed appropriate testing
methodology into their servers and their development practices
in general from the get-go, the savings would be as you suggest.
The savings would probably even have manifested during the
initial design and release, because I agree that building a
system with proper testing built-in does indeed make it possible
to save enormous amounts of bug-hunting later.

I also have little doubt (though more than none) that EQ's
original design and development did NOT include such tools.
Even though they may have been known practice as far back as,
oh, I think you've claimed 1995, they were hardly universal
then, let alone today. Lots of systems are designed without
taking the time to include any way to do the sorts of testing
that you keep harping on. Those systems are instead tested by
"trying them out and observing the results". Usually, such tests
involve fairly controlled cases where the expected results are
easy to verify. Those tests rarely cover all the possible cases
of a large and complex system. That means bugs go undetected
until the user base finds them. That's bad, but it happens.
It doesn't mean there was "no testing". It may mean there was
shoddy or very limited testing, but to keep claiming there is
"no testing" just means you've got a very narrow view of what
"testing" means.

None of us are claiming that Sony adequately tests their code.
All we're claiming is that they do SOME amount of something that
MOST of us are prepared to include under the broad term "testing".

Oh, and the other thing about building test tools into the design
and implementation of a system? It's a HECK of a lot more efficient
to do it from the start. Going back to a piece of software that's
evolved for nearly a decade under a shifting design team, and trying
to ADD such testing methodology after the fact, is a LOT harder.
Sure, it might eventually pay off before the game dies of old age.
But I don't really fault them for not putting in the effort to do it.

Post by the wharf rat
They don't test. "Testing" does not mean having your 7 year old
cousin run his twinked froglok wizard around PoK for 20 minutes before
you roll out the patch. Great Ghu, you can take a graduate degree in
Quality Control, did you know that? It's a regular *science*. And it's
*not* programming. It's more like applied statistics...

I agree, that's not what "testing" means. But frankly, having a
7-year-old run around PoK for 20 minutes IS one (very limited) form
of testing. If they did that, I would not say that they did "no"
testing. I'd say they didn't do nearly ENOUGH testing. And indeed,
though I'm sure they do more than 20 minutes of kiddie rampage, I
still don't think they do ENOUGH testing. But whatever they do, I
wouldn't call it "no testing". But you apparently would. Fine.
Enjoy your personal lexicon. Can we discuss something else now?

-- Don.

---------------------------------------------------------------------
-- See the a.g.e/EQ1 FAQ at http://www.iCynic.com/~don/EQ/age.faq.htm
--
-- Sukrasisx, Monk 68 on E. Marr Note: If you reply by mail,
-- Terrwini, Druid 58 on E. Marr I'll get to it sooner if you
-- Teviron, Knight 58 on E. Marr remove the "hyphen n s"
-- Wizbeau, Wizard 36 on E. Marr

[the wharf rat]

Post by Don Woods
I also have little doubt (though more than none) that EQ's
original design and development did NOT include such tools.

100% correct. IMHO what you do in that case is start adding
things, require developers to add unit test to the makefiles, add
standard instrumentation, etc. Build a test framework, nothing much at
first but open enough to be flexible. Keep a list of bugs that were
fixed: you had to know had to reproduce it, right? That's the basis
of your test. And you had to know how you knew it'd been fixed: there's
the expected output. It's an ongoing iterative process.

Taking the AA thing as an example: first, how will we know
our changes work? We need some kind of baseline. Ok, here's a program
that will call some of the important unit functions and here's its expected
output. Now, let's make our changes. Whoops, that's not the output we
expect... What's wrong? (*That's* testing by the way.)

Why is this effort less expensive than simply crossing your fingers
and letting your users find the problems? First of all it's reusable. Next
time you make an AA change you've already got at least the basic unit test.
Secondly, think of the effort that goes into using the let-your-customers-do-
it method. You actually do the patch. Your tech support guys take calls
until someone realizes there's a pattern. They follow whatever escalation
procedure exists. Some days later the problem gets to a developer, who has
to stop whatever she's doing and go figure out whether this is really a
problem and if so how to fix it. Then she fixes it (we hope! remember,
we really don't have any way of verifying that this time either...) and...
We start all over, with rolling out the patch and taking tech support calls.
And worst of all, we follow that procedure *every* time, and if our business
grows so does the expense of providing the support to your users while they
find your bugs. And lastly, what if your users find bug #1 today, you fix
it and repatch, and then they find bug #2 (sound familiar?) ?

So, IMHO refusing to implement sound development and quality
procedures because they're not there already is a mistake, as is refusing
to implement them because of the initial cost. Not only does a proper
framework cut the actual operating costs but it moves it from a variable
to a fixed cost (making your CFO very much happy). That lower cost
translates *directly* into a better game, because more cycles are available
for good stuff. And *that's* why this whole screw-it-the-players-will-find-
the-bugs-lets-go-play-foosball thing bothers me so very much.

[Richard Carpenter]

Post by the wharf rat

Post by Richard Carpenter

Post by the wharf rat

Post by Richard Carpenter
If such practices were common, you would see client crashes
resulting, rather than mere malfunctioning features.

That's a non sequitur.

Um, no, it's not. That most certainly *is* the sort of thing that is
very likely to happen when code is manipulated without the least bit
of testing.

Joining two clauses with a comma does not make them logically
connected. Lack of testing means that defects will appear in
production. It doesn't predict what those defects will be or where
they will appear.
To do that you'd have to TEST.

Of course it doesn't. I was merely illustrating the fact that they do
not release software with *no* testing. Maybe it's just splitting hairs,
but you are claiming exactly that over and over here.

I completely agree that the level of testing they *do* perform leaves
much to be desired, but let's not get ridiculous with our claims in
complaint.

Post by the wharf rat

Post by Richard Carpenter
I know of which I speak. If nothing else, programmers will run
the code as a safeguard against simple syntax errors.

Yup. Exactly the attitude that's getting Sony in trouble.
"It compiles. Ship it." They're not the first to take that attitude
and they won't be the last, but it's wrong on many levels.

Again, you're making wild assumptions.

Post by the wharf rat
BTW, we *do* see client crashes. I can crash my client simply
by leaving my trader up for a long enough time then trying to switch
characters. I can also crash it by switching back to the character
selection screen "enough". It's reproducible enough so that I always
log back to the desktop before doing anything important. That's not
the kind of bug I'm talking about. I'm talking about rolling out
production releases that are obviously broken in trivially TESTABLE
ways. All new AAs don't work. No songs work. Spell gems don't
refresh unless you zone.

I haven't seen a rash of people crashing due to AA bugs, and that's what
we're talking about here.

Post by the wharf rat

Post by Richard Carpenter

Post by the wharf rat
Ok, if they tested them why didn't they notice it didn't work?

Who knows? Certainly neither you nor I.

I do. They didn't notice because they didn't look. Because
if they HAD looked there's NO WAY they could have missed it.

You can't possibly know that, yet you continue to debate as if you do.
Like I've said, I'm no SoE apologist. I will complain about their poor
practices as much as the next guy. What I *won't* do, however, is
blatantly stretch the known truth to bolster my views.

Post by the wharf rat

Post by Richard Carpenter
What is much more certain,
however, is that they don't just throw in redesigns of such magnitude
without *any* testing. Not a chance.

Oh, think so?

What we've seen is most definitely proof of poor testing. Jumping to the
conclusion that it proves *no* testing is the true non sequitur.

Post by the wharf rat

Post by Richard Carpenter
Maybe they just assumed in that case that all was working.

Yes, exactly. They don't bother testing because it's just
a game, testing sucks, and anyway it's time for foosball. It's still
1991 at SOE.

More wild assumptions of little value to the discussion.

Post by the wharf rat

Post by Richard Carpenter
You're applying pie in the sky requirements to a shoestring budget,

Bullshit. Proper QA saves tons of money. It costs 100 times
as much to fix a bug that's been rolled out to production as it does
to fix it in system test, and it costs 10 times as much to fix it in
system test as it does to fix it in unit test. If someone at Sony
demanded they stop and get this right you'd all get your 25 new zones
hardly anyone will ever see each and every year and the rest of us
might not have to whine about the way hill giants warp when you try
and kite them.

I don't dispute the merits of QA. However, if the resources aren't
there, they just aren't there. They try to find a balance between
potential customer dissatisfaction and development costs. At this stage
of the game, do you really think they see the need for footing the bill
for the same level of QA and development procedures you might see in a
mission-critical enterprise application or world banking transaction
system? The practices you're demanding have their place, but I really
find it hard to justify them in an aging MMORPG with a dwindling player
base. They're striking a balance here. Simple as that.

Post by the wharf rat

Post by Richard Carpenter
Bottom line: anyone who says they implement game redesigns without any
testing is either loopy or just being a malcontent.

Got some evidence of that statement, you loopy malcontent who
doesn't seem to understand the first thing about proper software
engineering? (You're right! Gratuitous insults are fun! And here I
was wasting my time trying to keep this on an adult level...)

I understand it quite well, thank you. I also understand that the burden
of proving that they do not test changes at all is squarely on *you*. I
think it's you who are a bit unrealistic with your lofty standards under
the circumstances. Face facts. It's not worth it to SoE to put the sort
of resources toward this game that you demand. If it were, they would.

And if your inability or stubborn refusal to consider anything other
than simple negligence ("that's the way it is, because I say so!!!") as
the root of the problem is supposed to represent your efforts to "keep
this on an adult level", then you're not trying very hard.

[the wharf rat]

Post by Richard Carpenter
Of course it doesn't. I was merely illustrating the fact that they do
not release software with *no* testing. Maybe it's just splitting hairs,
but you are claiming exactly that over and over here.

Compiling it to check for syntax errors isn't testing.

Post by Richard Carpenter
Again, you're making wild assumptions.

Actually, I'm advertising the results of an informal analysis
based on publically available data and my own considerable expertise,
but argument from authority is particularly gauche when you cite yourself
so I've avoided the temptation.

Post by Richard Carpenter
I haven't seen a rash of people crashing due to AA bugs, and that's what
we're talking about here.

If client crashes are immaterial why did you bring them up?

Post by Richard Carpenter
You can't possibly know that, yet you continue to debate as if you do.

Of course I can. It's demonstrably true by observation. There's
no way anyone who'd checked to make sure that the new AA code was working
properly could have missed the fact that it wasn't. Therefore, they did not
look.

Unless you're willing to postulate that they looked, saw it was
broken, and just went ahead and released it anyway because what the hell
it's only a game?

Look, I love EQ. But watching these guys work is like sitting
in your favorite diner and watching them try and make hash browns
without boiling the potatoes first. You want to just walk over and
shake them: "Boil the spuds!! You can't make good hash browns unless you
boil the spuds!! Where'd you learn how to cook, anyway??"

Post by Richard Carpenter
I don't dispute the merits of QA. However, if the resources aren't
there, they just aren't there.

That's the same mistake a lot of managers make. The fact is
that an effective QA program saves much more than it costs, so arguing
against implementation on costs grounds is illogical. Any cursory
literature search will show this to be true.

Post by Richard Carpenter
They try to find a balance between
potential customer dissatisfaction and development costs.

I think they ignore the customers, since they realize that
it would take a series of major catatastrophes resulting in weeks
long outages and loss of XP/items to cause any kind of exodus. The
issue isn't customer satisfaction. It's quality engineering. They're
related but not necessarily causally. What *is* causally related to
proper QA is the bottom line on your 8K.

So they could make more money and have happier customers, but they
still don't bother. Is that dumb, or what?

Post by Richard Carpenter
And if your inability or stubborn refusal to consider anything other

"Nyahhh nyahh, if you don't admit I'm right you're not trying!"

Give me a break, son. With all your connections at Sony you ought
to be able to drag an engineer in here to describe their development
process. Instead of asking me to accept that a successful make counts
as a quality assurance program.

[Don Woods]

Post by the wharf rat
Of course I can. It's demonstrably true by observation. There's
no way anyone who'd checked to make sure that the new AA code was working
properly could have missed the fact that it wasn't. Therefore, they did not
look.

Sigh, I've been avoiding getting involved in this flame-fest, and
by all sanity should probably continue to stay out of it, but I'm
getting really tired of the Rat saying the same crap over and over.

I don't know if the Rat really has any direct experience with
software development practices, but I do. I've been a professional
software developer for over 30 years. I know EXACTLY how EASY it
is for even a good QA team to overlook problems that are glaringly
obvious once someone points them out. (And I'm not claiming SOE's
QA team is anywhere near "good".)

Now, I'll grant that the tale of Bards Who Can't Sing is impressive.
If it really was the case that no bard could sing any song at all,
and they missed it, that's hard to explain short of "all we're making
is a trivial change, no need to test much", which is shabby practice
at best. But this AA thing? Sure, I'll bet they had someone try it
out. That person probably took one or more characters with lots of
AAs (perhaps via testbuffing) and checked out the UI in the new AA
window, maybe even buying some AAs, verifying the AAs were displayed
right, whatever. Everything I've seen posted here seems to indicate
that the bugs had to do with the AAs not having the effects they're
supposed to. That is NOT going to be "obvious" to someone who's just
checking whether the AA window shows the AAs to be present. And if
existing AA hotkeys stopped working, but new hotkeys worked, I could
see that getting overlooked in testing also. Shabby practice again,
but easy to overlook.

I'll well believe that they didn't take any characters with lots of
AAs and run them around to check if the AAs were working. Even if
they had, I don't think everyone is as good as the Rat at noticing
that they're not getting the expected 30% bonus to critical hit
chances, or whatever else the AAs were supposed to do. Indeed, given
how hard it is to test each individual AA, I could easily believe that
the AAs themselves were not tested much if at all. But that's a FAR
CRY from "no testing". I'm sure they tested the UI itself. All that
some of us are trying to say is that screaming "SOE never does any
testing at all" is an obvious hyperbole that weakens your argument.

Now if you just want to bitch about all the bugs that they overlooked
during their testing, fine. If you want to bitch about bugs that you
know were reported by players on the Test server (especially if you
know because you reported them yourself) and that got left in when the
code went live, fine. We've all seen examples of this, and we all
agree that SOE's QA process is crappy. I'm willing to speculate that
they don't have much if any instrumentation to enable automatic
testing; for instance, no, they probably CAN'T get into a combat and
then look at a log to verify things like critical hits. I'll bet they
just log into the game and try some things, sure with extra abilities
like testbuffing or invulnerability or whatever so they can set up
specific tests when appropriate, but it's probably NOT automated and
they can't possibly manually test ALL the features.

The sad truth is that when you have hundreds of thousands of users,
those users are going to do a MUCH better job of finding the bugs than
any QA team. So you test some stuff and then hope for the best. No,
it's not good practice, and they could certainly do a better job even
with the resources they do have, but I'm sure they do SOME testing.

-- Don.

---------------------------------------------------------------------
-- See the a.g.e/EQ1 FAQ at http://www.iCynic.com/~don/EQ/age.faq.htm
--
-- Sukrasisx, Monk 68 on E. Marr Note: If you reply by mail,
-- Terrwini, Druid 58 on E. Marr I'll get to it sooner if you
-- Teviron, Knight 58 on E. Marr remove the "hyphen n s"
-- Wizbeau, Wizard 36 on E. Marr

[the wharf rat]

Post by Don Woods
That person probably took one or more characters with lots of
AAs (perhaps via testbuffing) and checked out the UI in the new AA
window, maybe even buying some AAs, verifying the AAs were displayed
right, whatever.

Having someone click through a few paths in the UI is NOT
testing, much less QA. First of all, they changed the AA system, right?
So they should test to make sure that the AA system works. What are you
testing - and I glorify it by the use of the word - when you click through
a few spots and see if things look like they're displayed correctly?

The UI. Not the AA system. Testing the AA system would require
something like, oh, I dunno, *instrumenting* *the* *code* to output the
results of AA calculations. Then calling the routines outside of the
actual game to verify them.

Post by Don Woods
I'll well believe that they didn't take any characters with lots of
AAs and run them around to check if the AAs were working. Even if

That's not testing. Charitably, I might go so far as classifying that
as user acceptance. QA is not some half-assed we-clicked-it-and-it-looked
OK nonsense. It's a scientific approach to software verification that's
aimed at reducing or even elimininating defects that make it into the
released product. I'm not sure why people who claim to be familiar with
the software development process seem to have such a problem with this.

Post by Don Woods
they had, I don't think everyone is as good as the Rat at noticing
that they're not getting the expected 30% bonus to critical hit

That's why you automate the process. Supply known input and
examine the output for correctness. You don't hire your teenaged cousin
to run a few characters around and make sure the client doesn't crash.

Post by Don Woods
The sad truth is that when you have hundreds of thousands of users,
those users are going to do a MUCH better job of finding the bugs than
any QA team. So you test some stuff and then hope for the best.

Oh, bwahaha, /snort, stop it, you're killing me, bwahaha...

Great Ghu, did you *really* just say that there's no point to
testing software because your users will find the bugs??? Donald
Knuth is turning over in his grave... I'm glad the people who write
ATM software don't feel that way. Even LINUX people test their code,
and not by booting a 486 and declaring victory if they get a console prompt.


"If builders built buildings the way programmers
wrote programs, then the first woodpecker that
came along would destroy civilisation."

[Schadenfreude]

Post by the wharf rat
Oh, bwahaha, /snort, stop it, you're killing me, bwahaha...
Great Ghu, did you *really* just say that there's no point to
testing software because your users will find the bugs??? Donald
Knuth is turning over in his grave... I'm glad the people who write
ATM software don't feel that way. Even LINUX people test their code,
and not by booting a 486 and declaring victory if they get a console prompt.

2005

Toyoya Prius

Last month automaker Toyota announced a recall of 160,000 of its Prius
hybrid vehicles following reports of vehicle warning lights
illuminating for no reason, and cars' gasoline engines stalling
unexpectedly. But unlike the large-scale auto recalls of years past,
the root of the Prius issue wasn't a hardware problem -- it was a
programming error in the smart car's embedded code. The Prius had a
software bug.

1985-1987

Therac-25 medical accelerator.

A radiation therapy device malfunctions and delivers lethal radiation
doses at several medical facilities. Based upon a previous design, the
Therac-25 was an "improved" therapy system that could deliver two
different kinds of radiation: either a low-power electron beam (beta
particles) or X-rays. The Therac-25's X-rays were generated by
smashing high-power electrons into a metal target positioned between
the electron gun and the patient. A second "improvement" was the
replacement of the older Therac-20's electromechanical safety
interlocks with software control, a decision made because software was
perceived to be more reliable.

What engineers didn't know was that both the 20 and the 25 were built
upon an operating system that had been kludged together by a
programmer with no formal training. Because of a subtle bug called a
"race condition," a quick-fingered typist could accidentally configure
the Therac-25 so the electron beam would fire in high-power mode but
with the metal X-ray target out of position. At least five patients
die; others are seriously injured.

November 2000

National Cancer Institute, Panama City.

In a series of accidents, therapy planning software created by
Multidata Systems International, a U.S. firm, miscalculates the proper
dosage of radiation for patients undergoing radiation therapy.

Multidata's software allows a radiation therapist to draw on a
computer screen the placement of metal shields called "blocks"
designed to protect healthy tissue from the radiation. But the
software will only allow technicians to use four shielding blocks, and
the Panamanian doctors wish to use five.

The doctors discover that they can trick the software by drawing all
five blocks as a single large block with a hole in the middle. What
the doctors don't realize is that the Multidata software gives
different answers in this configuration depending on how the hole is
drawn: draw it in one direction and the correct dose is calculated,
draw in another direction and the software recommends twice the
necessary exposure.

At least eight patients die, while another 20 receive overdoses likely
to cause significant health problems. The physicians, who were legally
required to double-check the computer's calculations by hand, are
indicted for murder.

Pretty much anything software driven can crash; to use your quoted
example of ATMs there are numerous examples of BSoD, the error when
Germany transferred from the DM to the Euro, the US Fujitsu disaster
and countless others.

"Security researchers uncovered nearly 5,200 software vulnerabilities
in 2005, almost 40 percent more than the number discovered in 2004,
according to Washingtonpost.com. From the article: 'According to
US-CERT...researchers found 812 flaws in the Windows operating system,
2,328 problems in various versions of the Unix/Linux operating systems
(Mac included). An additional 2,058 flaws affected multiple operating
systems.'"

Even new MMORPGs like WoW, Vanguard and Conan are incredibly bug
ridden nevermind ones closing in on their tenth birthday where no one
currently in charge of maintaining the code was around when much of it
was written.

[murdocj]

On Sun, 05 Aug 2007 11:09:45 +0100, Schadenfreude

Post by Schadenfreude

Post by the wharf rat
Oh, bwahaha, /snort, stop it, you're killing me, bwahaha...
Great Ghu, did you *really* just say that there's no point to
testing software because your users will find the bugs??? Donald
Knuth is turning over in his grave... I'm glad the people who write
ATM software don't feel that way. Even LINUX people test their code,
and not by booting a 486 and declaring victory if they get a console prompt.

2005
Toyoya Prius
... list of bugs...
Even new MMORPGs like WoW, Vanguard and Conan are incredibly bug
ridden nevermind ones closing in on their tenth birthday where no one
currently in charge of maintaining the code was around when much of it
was written.

I can tell you from experience that WoW has *far* fewer bugs than EQ.
Vanguard, on the other hand, is reputed to be incredibly bug ridden
(and that's based on reports from people who have played a wide
variety of MMOs, including EQ, EQ2, and WoW). So the argument that
"all MMOs have a ton of bugs" just doesn't make it.

Yes, testinng is hard. But as the "wharf rat" pointed out, there are
things you can do, like regression testing the new AA calculations
against the old, that aren't rocket science and don't require checking
every path through the code. They just require a mindset that you are
going to test what can be tested before you release the code to
customers. And over the long haul, that would save work for the EQ
development team.

[Schadenfreude]

On Sun, 05 Aug 2007 12:31:14 GMT, murdocj

Post by murdocj
I can tell you from experience that WoW has *far* fewer bugs than EQ.
Vanguard, on the other hand, is reputed to be incredibly bug ridden
(and that's based on reports from people who have played a wide
variety of MMOs, including EQ, EQ2, and WoW). So the argument that
"all MMOs have a ton of bugs" just doesn't make it.

I've played ALL the games mentioned in either late beta or at release
and all had (and continued to have up until I stopped playing them)
numerous bugs.

In fact I cannot think of any PC game which didn't have bugs and
needed patching.

[murdocj]

On Sun, 05 Aug 2007 14:18:59 +0100, Schadenfreude

Post by Schadenfreude
On Sun, 05 Aug 2007 12:31:14 GMT, murdocj

Post by murdocj
I can tell you from experience that WoW has *far* fewer bugs than EQ.
Vanguard, on the other hand, is reputed to be incredibly bug ridden
(and that's based on reports from people who have played a wide
variety of MMOs, including EQ, EQ2, and WoW). So the argument that
"all MMOs have a ton of bugs" just doesn't make it.

I've played ALL the games mentioned in either late beta or at release
and all had (and continued to have up until I stopped playing them)
numerous bugs.
In fact I cannot think of any PC game which didn't have bugs and
needed patching.

Yes, all game programs have bugs.

No, they do NOT have the same magnitude of bugs. Some games are
buggier than others. That is such a truism that it amazes me that I
even have to state it, or that anyone would argue about it.

When I played EQ, it was not unusual to run into bugs. In WoW, I
rarely run into bugs. Apparently in Vanguard, at least upon release,
the bugs were so numerous that it required an act of will to keep
playing despite them.

[Schadenfreude]

On Sun, 05 Aug 2007 16:46:36 GMT, murdocj

Post by murdocj
Yes, all game programs have bugs.
No, they do NOT have the same magnitude of bugs. Some games are
buggier than others. That is such a truism that it amazes me that I
even have to state it, or that anyone would argue about it.
When I played EQ, it was not unusual to run into bugs. In WoW, I
rarely run into bugs. Apparently in Vanguard, at least upon release,
the bugs were so numerous that it required an act of will to keep
playing despite them.

SWG had more bugs than WoW on release, AO had more bugs than WoW on
release, EQ had more bugs than WoW on release, Vanguard had more bugs
than WoW on release. None of these other games had remotely the
resource behind them that WoW did. It's not rocket science.

Vanguard had 1 [ONE] person devoted to QA so it's hardly surprising
the enormous resources behind WoW were able to do better HOWEVER
unless you're some sort of rabid fanboy or didn't actually PLAY WoW on
release then to claim it was even remotely bug free is purely and
simply delusional.

Of course what would I know I only actually PLAYED all these games at
release, eh? I must have imagined the showstopping loot lag, mob
warping, harvesting freeze bugs and mysterious-without-warning
disconnections every 20 minutes in the first week of WoW. Not to
mention the "oops, seems you don't HAVE any characters on that server"
and all the other fun stuff even getting logged in. Hell, it was more
stable the last week of Beta.

All this in a very much 2nd generation game.

[murdocj]

On Mon, 06 Aug 2007 01:17:25 +0100, Schadenfreude

Post by Schadenfreude
On Sun, 05 Aug 2007 16:46:36 GMT, murdocj

Post by murdocj
Yes, all game programs have bugs.
No, they do NOT have the same magnitude of bugs. Some games are
buggier than others. That is such a truism that it amazes me that I
even have to state it, or that anyone would argue about it.
When I played EQ, it was not unusual to run into bugs. In WoW, I
rarely run into bugs. Apparently in Vanguard, at least upon release,
the bugs were so numerous that it required an act of will to keep
playing despite them.

SWG had more bugs than WoW on release, AO had more bugs than WoW on
release, EQ had more bugs than WoW on release, Vanguard had more bugs
than WoW on release. None of these other games had remotely the
resource behind them that WoW did. It's not rocket science.
Vanguard had 1 [ONE] person devoted to QA so it's hardly surprising
the enormous resources behind WoW were able to do better HOWEVER
unless you're some sort of rabid fanboy or didn't actually PLAY WoW on
release then to claim it was even remotely bug free is purely and
simply delusional.
Of course what would I know I only actually PLAYED all these games at
release, eh? I must have imagined the showstopping loot lag, mob
warping, harvesting freeze bugs and mysterious-without-warning
disconnections every 20 minutes in the first week of WoW. Not to
mention the "oops, seems you don't HAVE any characters on that server"
and all the other fun stuff even getting logged in. Hell, it was more
stable the last week of Beta.
All this in a very much 2nd generation game.

Going back to the original point, and not worrying about whether WoW
in particular is less or more buggy than other games... the argument
that "all games have bugs, therefore don't complain about them" is
nonsense. Yes all games have bugs, no all games do NOT have the same
magnitude of bugs. You yourself cite reasons why different games have
different numbers of bugs.

As far as having played a lot of games... I played EQ for 6 years, WoW
since a few weeks after release, EQ2 for about 4 months, Guild Wars
for a few months, LOTRO beta, a bit of Auto Assault beta, etc. My
wife has played all of the above plus SWG (more hours than I want to
think about) a variety of other MMOs and MUDs going back to 300 baud
dialup modems. When she says a game is buggy, I trust her. So please
don't imagine that you're going to impress me with your vast MMO
experience.

[the wharf rat]

Post by Richard Carpenter
Oh come on. Do you really believe that they roll out major fixes without
any testing?

Yes.

Any half-assed automatic testing program would have revealed
that the new AA calculations were broken. The test harness calls the
AA calculation routines and compares the output to the expected. That
wasn't done, or they'd have known right away it was broken.

I understand how subtle playability bugs or complex interactions
between game events and player input aren't necessarily testable, but
"level 10 AA should give +30% to skill foo. Does it?" is an easy condition
to test for.

Post by Richard Carpenter
That's absurd.

Yes, it is. That trivially verifiable conditions aren't even
checked before rollout is simple bad practice. It's like the stupid
guild hall door bug. That should never reappear because it should be
part of the automatic test harness every rev goes through as part of
routine QA. Of course, there IS no QA at Sony.

Look, after 9 years these guys should have an automatic test suite
1 million lines long. Every time they find a bug like the door bug there
should be a new test case added - "Input click to start-zoning-function.
Is zoning routine called?" - but they don't do that, or trivial bugs wouldn't
reappear. This isn't a black art. How do you think they verify Oracle,
for instance?

Post by Richard Carpenter
For every change that is so lamented in the
forums for it's perceived haphazard testing and rollout, I would
honestly bet that there are a hundred that go in without any trouble at
all.

They "go in without any trouble at all" by accident, rather than
because they were verified. Maybe we're in a semantic conflict. "Software
testing" doesn't mean exhaustively verifying every possible combination of
inputs and outputs. It means running the program through a test harness
that contains test cases for combinations it's possible to test for,
test cases for known bugs that should remain fixed, and some "corner
cases" - extreme combinations. If a developer finds and fixes a stuck
in the geometry bug she should create a test case that verifies that bug
is fixed and add it to the test harness. And etc.

The result is a constant if slow improvement in software quality.
As opposed to "well the door's broken again good thing we left in the
workaround to the patch to the fix so we can ignore it".

[stanmann]

Post by the wharf rat
Look, after 9 years these guys should have an automatic test suite
1 million lines long. Every time they find a bug like the door bug there
should be a new test case added - "Input click to start-zoning-function.
Is zoning routine called?" - but they don't do that, or trivial bugs wouldn't
reappear. This isn't a black art. How do you think they verify Oracle,
for instance?

Uh, firstly automated test suites aren't for testing trivial bugs.
they are for testing complicated boundary conditions that can't be
reached in normal test conditions.

Second, the original dev team didn't do much of anything right, so an
automated test suite is right out.

Hacking an automated test suite together for someone elses spagetti
code that mostly works is quite frankly a waste of time, and I'd
rather they be working on new content, thank you very much

--
StanMann

[murdocj]

Post by stanmann

Post by the wharf rat
Look, after 9 years these guys should have an automatic test suite
1 million lines long. Every time they find a bug like the door bug there
should be a new test case added - "Input click to start-zoning-function.
Is zoning routine called?" - but they don't do that, or trivial bugs wouldn't
reappear. This isn't a black art. How do you think they verify Oracle,
for instance?

Uh, firstly automated test suites aren't for testing trivial bugs.
they are for testing complicated boundary conditions that can't be
reached in normal test conditions.

Nonsense. Automated test suites are for automating as much as
possible of the testing. The last thing you want to do is only test
weird conditions... you want to test normal cases, error conditions,
etc etc etc. Basically you want to catch as many errors as early as
is possible.

Post by stanmann
Second, the original dev team didn't do much of anything right, so an
automated test suite is right out.
Hacking an automated test suite together for someone elses spagetti
code that mostly works is quite frankly a waste of time, and I'd
rather they be working on new content, thank you very much

Obviously if there's no automated testing, they can't introduce tests
for everything. But what they can do is introduce testing *as they
introduce changes*. So when you change how AAs work, you introduce
tests that verify that the old AAs and the new AAs work the same.

[Richard Carpenter]

Post by the wharf rat

Post by Richard Carpenter
Oh come on. Do you really believe that they roll out major fixes
without any testing?

Yes.
Any half-assed automatic testing program would have revealed
that the new AA calculations were broken. The test harness calls the
AA calculation routines and compares the output to the expected. That
wasn't done, or they'd have known right away it was broken.
I understand how subtle playability bugs or complex interactions
between game events and player input aren't necessarily testable, but
"level 10 AA should give +30% to skill foo. Does it?" is an easy
condition to test for.

Post by Richard Carpenter
That's absurd.

Yes, it is. That trivially verifiable conditions aren't even
checked before rollout is simple bad practice. It's like the stupid
guild hall door bug. That should never reappear because it should be
part of the automatic test harness every rev goes through as part of
routine QA. Of course, there IS no QA at Sony.
Look, after 9 years these guys should have an automatic test suite
1 million lines long. Every time they find a bug like the door bug
there should be a new test case added - "Input click to
start-zoning-function. Is zoning routine called?" - but they don't do
that, or trivial bugs wouldn't reappear. This isn't a black art.
How do you think they verify Oracle, for instance?

Post by Richard Carpenter
For every change that is so lamented in the
forums for it's perceived haphazard testing and rollout, I would
honestly bet that there are a hundred that go in without any trouble
at all.

They "go in without any trouble at all" by accident, rather than
because they were verified. Maybe we're in a semantic conflict.
"Software testing" doesn't mean exhaustively verifying every possible
combination of inputs and outputs. It means running the program
through a test harness that contains test cases for combinations it's
possible to test for, test cases for known bugs that should remain
fixed, and some "corner cases" - extreme combinations. If a developer
finds and fixes a stuck in the geometry bug she should create a test
case that verifies that bug is fixed and add it to the test harness.
And etc.
The result is a constant if slow improvement in software quality.
As opposed to "well the door's broken again good thing we left in the
workaround to the patch to the fix so we can ignore it".

I would imagine it's a bit more complicated than you claim to implement
a "test harness" that would put a character though various mob
encounters, triggering all of the spells and abilities at their
disposal. I'm not saying it can't be done. I'm just saying it's well
within the realm of possibility that their testing mechanisms merely
were not updated to consider the problem at hand.

Geez. Is it so hard to fathom how something just may have slipped
through the cracks? To complain with such zeal about one bug in such a
huge project is pretty unreasonable, especially considering the fact
that it was turned around pretty quickly.

Show me this perfect development team you've envisioned that is
magically able to catch any and all bugs in their enormous multi-server
systems that are accessed on a world-wide scale, and accomplish this
with a tiny fraction of the number of people and resources such an
endeavor would require.

[the wharf rat]

Post by Richard Carpenter
I would imagine it's a bit more complicated than you claim to implement
a "test harness" that would put a character though various mob
encounters, triggering all of the spells and abilities at their
disposal.

"Software testing" doesn't mean exhaustively verifying every possible
combination of inputs and outputs. Even when your target involves a very
complicated graphical UI it's not necessary to test for every possible
combination of keystroke and mouse click. Every program is a finite state
machine ***or can be simplifed to approximate a finite set of states for
the purpose of verification***. For instance, to verify proper interaction
between a moving character and the geometry you don't need to test every
possible path. A path down each axis and a path into two corners is probably
98% sufficient. And the beauty is that even if the ONLY way to test this is
to actually have a human drive a GUI tools exist that will "record" that
session and produce an automated test that can be run by a machine! Then
EVERY TIME someone touches the code a machine automatically makes sure that
there's at least a 98% chance that you can still move around correctly.

There's much more at stake than downtime or flawless patches. Whether
you believe (as I do) that these guys should spend some time fixing long-
standing playability issues like warping mobs or if you feel that they should
put out as much new content as possible there are a finite number of
resources available. A bug that makes it to production is orders of
magnitude (yes, orders of magnitude) more expensive to fix than a bug
caught in unit test. If Sony has 200 programmer units to use, they can
use them to fix 2 bugs in the production system, or they can use 100 of them
for effective QA, 50 of them to fix 10 bugs during unit test, and 50
to design new zones only 2% of the playerbase will ever see. Which is better?

Think of effective, scientific software QA as Lesson of the Devoted
for defect prevention.

Post by Richard Carpenter
Geez. Is it so hard to fathom how something just may have slipped
through the cracks?

This isn't an isolated incident. It's so much a part of a set
pattern that people accept it as normal behavior. That in itself is
a serious defect (that it's part of a pattern, not that people accept it).

Post by Richard Carpenter
To complain with such zeal about one bug in such a

Pattern. I'm fruitlessly whining about Sony's observable lack of
adequate development and operational quality control. It's a consistent
pattern, not a single bug.

Post by Richard Carpenter
magically able to catch any and all bugs in their enormous multi-server

Not magic. Just standard scientific development procedures.
And not all of them, or course. But really, *every* time they change
something it's broken for N weeks afterward. That's not just bad luck.

[Richard Carpenter]

Post by the wharf rat
But really, *every* time they change
something it's broken for N weeks afterward. That's not just bad luck.

It's also blatantly untrue. That's the sort of bunk that I take issue with.
If people want to complain about bugs and/or playability issues, that's
fine. They pay for the right to do so. However, when that complaining is
done through blatant over-exaggeration and rhetoric, it becomes little more
than noise pollution, which gets old real quick.

[the wharf rat]

Post by D.J.
If the servers are down for patch, it will say so.

I actually don't care if they're down. I'd like to know
that they're going to be down, so, for instance, I don't get up
2 hours early to finish a quest before work and find out I have to camp
in a safe spot.

There's no reason not to put 24 hours notice in the MOTD. Even
if nice people on Usenet send me urls to otherwise-useless websites that
contain the network status information I expect to see in the page
prominently marked network status... It's sort of like if I went to
Home Depot and walked over to the plumbing section but all that was on the
shelf was a broken light bulb because I'm expected to know that to find
the plumbing section I need to go to the men's room and read the writing
on the third stall from the left.

[Richard Carpenter]

Post by the wharf rat

Post by D.J.
If the servers are down for patch, it will say so.

I actually don't care if they're down. I'd like to know
that they're going to be down, so, for instance, I don't get up
2 hours early to finish a quest before work and find out I have to
camp in a safe spot.
There's no reason not to put 24 hours notice in the MOTD. Even
if nice people on Usenet send me urls to otherwise-useless websites
that contain the network status information I expect to see in the
page prominently marked network status... It's sort of like if I went
to Home Depot and walked over to the plumbing section but all that was
on the shelf was a broken light bulb because I'm expected to know that
to find the plumbing section I need to go to the men's room and read
the writing on the third stall from the left.

Well, if that *did* happen, would you find a Home Depot employee and ask
them to help you, or would you stand there and "rage against the dying of
the light?" :P

[the wharf rat]

Post by Richard Carpenter
Well, if that *did* happen, would you find a Home Depot employee and ask
them to help you, or would you stand there and "rage against the dying of
the light?" :P

I suppose I could call Sony tech support once a day to find
out if they have a patch scheduled. As if they'd know. 1/2 :-)

[Richard Carpenter]

Post by the wharf rat

Post by Richard Carpenter
Well, if that *did* happen, would you find a Home Depot employee and
ask them to help you, or would you stand there and "rage against the
dying of the light?" :P

I suppose I could call Sony tech support once a day to find
out if they have a patch scheduled. As if they'd know. 1/2 :-)

Or merely listen to those around you and use that newfound knowledge to
solve your particular problem.

[Tony]

Post by the wharf rat
I actually don't care if they're down. I'd like to know
that they're going to be down, so, for instance, I don't get up
2 hours early to finish a quest before work and find out I have to camp
in a safe spot.

Well, this patch was an odd one, because it was announced at the last
minute on the boards. However, I still managed to provide plenty of
warning to my guild mates that it was due.

Previous patches have been more clearly advertised on the boards and
I've usually provided around 5-10 days notice to my guild mates of the
impending patch.

[Eric D. Braden]

From the SOE page on updateshttp://eqplayers.station.sony.com/game_updates.vm
We currently do not have information about the next Game Update.
All Updates
Upcoming Upcoming Updates
7/25/2007 July 25, 2007
Its 7/31, they've had this update scheduled since at least last night
because I heard people talking about it in general channel. And yet a)
they have no information about it and b) don't even know its going on!
Patch had been going for 30 minutes when I checked this.

Nothing restores activity to a quiet newsgroup like a good, old-
fashioned argument :)

[Lief]

Post by Eric D. Braden

Post by Lance Berg
From the SOE page on

updateshttp://eqplayers.station.sony.com/game_updates.vm

Post by Eric D. Braden

Post by Lance Berg
We currently do not have information about the next Game Update.
All Updates
Upcoming Upcoming Updates
7/25/2007 July 25, 2007
Its 7/31, they've had this update scheduled since at least last night
because I heard people talking about it in general channel. And yet a)
they have no information about it and b) don't even know its going on!
Patch had been going for 30 minutes when I checked this.

Nothing restores activity to a quiet newsgroup like a good, old-
fashioned argument :)

that was the plan :p