[OT] Fundamental Issues with Open Source Software Development

[Tim Wunder]

On 4/16/2004 7:54 AM, I believe that David A. Bandel wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> On Thu, 15 Apr 2004 21:51:09 -0500
> Alma J Wetzker <almaw at ieee.org> wrote:
> 
> 
>>Klaus-Peter Schrage wrote:
>>
>>>http://www.citizenlab.org/opensource/
>>
>>I have to wonder if the project has reached a 1.x release number.
>>
>>
>>>Canadian researcher Michelle Levesque has thoroughly examined an
>>>Open Source Project (which she kept anonymous) and found "the five
>>>most important flaws with Open Source software development to be as
>>>follows: 1) User interface design
>>
>>I worked on a project for a bit over a year.  I did the "internal"
>>feature stuff and there were four programmers working on the user
>>interface.  That seems to be about the correct balance between
>>features and interface in time and effort.
>>
>>
>>>2) Documentation
>>
>>Yeah, read the code.  ;)
> 
> 
> No, she's looking for a pretty book she won't read.
> 

Indeed. If the program does its job well enough, documentation is hardly 
required. FWIW, what documentation does MS provide for Windows? Their 
help documents? Gimme a break. Has anybody ever really been helped by 
them? You get better help from, you got it, the user community. And the 
new help system in MS Office is horrid, IMO.

> 
>>>3) Feature-centric development
>>
>>I like projects where you add the features you want, get them working 
>>correctly and move on.  I am just anal-retentive enough to make sure
>>they all work together in various combinations.  (expect is great.)
>>
>>
>>>4) Programming for the self
>>
>>How many programming products do you know that did not start as
>>someone needing something that they could not do with another program?
>> (Be that a 
>>feature, knowledge of a system or not enough money to buy the "real"
>>package.)
> 
> 
> So who else you gonna program for without pay?
> 

Since when does Open Source require programmers work without pay? That's 
largely the case, I'll grant you, but it is not a requirement. How many 
kernel developers are currently working on it without pay?

> 
>>>5) Religious blindness"
>>
>>FOSS programmers tend to be passionate about their projects.  That is
>>a sign of a good programmer, they care about the system and the uses. 
>>It seems to be more focus and less blindness.  Besides, Has anyone
>>heard a MCSE talk about linux v windoze lately?  There are idiots on
>>all sides, that has nothing to do with the merits of the position.
> 
> 
> The phrase "religious blindness" is used by zealots when you don't come
> to see how their viewpoint (religion) is the correct one.
> 
> 
>>>Has someone an idea what "Project X" could be? At first sight, I
>>>thought of the ALSA Project or The Gimp, but Project X seems to come
>>>with a calendar ...
>>>Klaus
>>
>>With the many FOSS products available, it is foolish to generalize too
>>much. Various programmers have various strengths.  As for copying an
>>interface directly, as the author suggests, that may violate legal as
>>well as ethical principles.  We are a maturing community, give us
>>time.
> 
> 
> Why copy garbage?  My first graphical interface (GUI) was mwm (Motif
> Window Manager).  The M$ windows interface (this goes for KDE and Gnome
> too) is unintuitive and fights me _way_ too much.  I still can't copy
> and paste from one window to another in Windows using just the mouse --

Highlight, right-click, choose copy; right-click, choose paste. Not 
nearly as nice as highlight to copy and middle-click to paste, but can 
be done with just the mouse...

> annoying.  And command completion and command history is a must.  Now, I

You got that. Windows command line is nearly useless.

> wouldn't know a good interface from a bad one, I just know what works
> (and doesn't work) for me.
> 

Indeed.

Regards,
Tim