Merchant Archive

Thread: Timeline for fixes

Haruspex77
Tue Jun 15, 2004 4:35 pm
#14






DocSavag wrote:

Its really not possible to manage a project like SWG without pretty detailed design documentation.




Sure it is, I have seen it done. Prior to the mid 1970's most development was done without significant documentation. The problem was the high rate of development project failure. In the early 70's more than half of all system development projects failed to result in anything going into service. All of the time and money spent was lost.


Success without all the overhead of documentation is possible if the right people are completely available. It doesn't take much distraction, though, for it to crumble into disaster. I know that SOE has some people with that kind of ability from what some have written, and from some of the bio's. But it is discouragingly easy for even the best people to fail, at least partially. Not the best approach in the long run.


I hope that SOE's development process is better than I fear, but their is discouragingly much evidence to keep my fears alive.


The kind of evidence that concerns me is SOE's inability to give answers to simple questions. The kind that should be able to be looked up in seconds with a copy of the documentation in hand. Instead they have come back with answers that are provably wrong in some cases, or outright admitted that they don't know. The best example is the "BE advanced guide" where there are explicit statments to that effect.


[One could claim that those statements were made by players, not devs, due to the extensive use of postings (by the very BE's that were requesting a guide) in producing it. That, however, ignores the fact that SOE did edit the document, did not attribute particular sections to the authors, and published the result under their imprimature. They may have been players' statments that "we don't know" originally, but they made them their own.]


Other fine examples are the armour guide, or the skill tapes information which were seriously wrong on many points in ways that were obvious to a knowledgable player. It is possible that much of this is due to a failure to maintain documentation after changes or discovery of "shippable" bugs. However, if the documents are not maintained as the project develops, that is almost the same as never writing any.


There have also been a number of cases where SOE would have clearly gotten a lot of milage out of revealing some of their plans but did not. "Don't know" statements by TH are common. And other more subtle things others have commented on about the structure of bugs all indicate a weakness in their development process. There are other things that could go wrong that could have caused that, like an accepted lack of accountabilityby programmers to read and follow the specifications.


The upside to this is that the recent team comments by Cinco Barnes specificly refers to writing design documentation! This is at least a hint that they recognize the need.

Haruspex77
Tue Jun 15, 2004 5:01 pm
#15






DocSavag wrote:


That assumes the ratio of houses to players (actively paying for service) is 1:1. It isn't.




Are you claiming to know that for a fact, which implies that you know what it is?


1:1 does not seem like a bad estimate to me for the estimate of the characters to houses ratio. Lots of characters don't have houses, especially little used alts. Some have many (up to 10 small Naboo's). Just how it averages out is unknown to me but 1:1 is in the middle of the likely range.


Note that from the recent survey that there are more than 1.5 accounts per player,at least for forum participants, and each account can have several characters on various servers + Jedi.


I would guess at about 100k active accounts, or $1.5 million a month. Even with equipment costs, customer service,GS&A, etc. that should support a staff of 50 developers if they aren't pullingcashflow out of the project rather than investing. For personal reasons, I suppose would rather that they invested in some of the rising bands on the Epic lable, the Mouse is roaring!




DocSavag
Wed Jun 16, 2004 12:00 am
#16

We can discuss theory or we can discuss practice but I maintain that very few design documents would include a detail as mundane as "why" there are only 6 vendors instead of 7. The answer might well be because no on thought about it.





----------------------------------
Chataka Windae
Rifleman/Combat Medic
CEO, Windae Enterprises
Mesric Sanctuary Founder



p4Samwise
Wed Jun 16, 2004 12:06 am
#17






DocSavag wrote:


The answer might well be because no on thought about it.




I agree, that's the most likely explanation. Clear evidence of arobustdevelopment process at work.




"Prettiest shim on Bria!" - Sev
Certified "cool" by the Darth Vader of Bria

Blue glowie.
Page 2 of 2