httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Roy T. Fielding" <>
Subject Re: Front end needs some work
Date Fri, 02 Jul 1999 00:50:38 GMT
>> Apache provides a unique opportunity for people to change 
>> something if they think they know a better way. 
>Yes, it does. So does any other GNU project. On a certain level

Apache is not a GNU project.

>this 'our code and you are welcome to it' thing is great. You take 
>the wheat with the chaffe and get out your sickle blade if you want. 
>The choice is yours ( if you have the talent to mess with it ).
>I 'change' Apache all the time. I make it do things it currently 
>can't do. That's all well and good... but 'changing something' 
>and then having it end back in an official 'release' of the software
>is something else entirely. Apache is a 'meritocracy'... there
>is a small core of people at the top who are 'in charge' and they
>also are the ones who must 'approve' everything. Of course it
>must be this way or there would be chaos... but if you have 
>ever worked for a Fortune 500 company you would understand
>that in this respect Apache is no different from any other
>'company'. Ideas must float to 'the top' and just because an
>idea is a 'good' one doesn't mean it will make an appearance
>in the company's product line. I have seen many a 'good' idea
>get rejected at Apache in the same way they do at the places
>where I been employed. Sometimes it's just good 'ol politics.

When presented with a set of competing alternatives, it is necessary to
look at all aspects of the solutions in order to determine which is best,
and, yes, sometimes that includes determining where the solution is coming
from and how that will impact its long-term maintainability.  Whether you
call this systems engineering, social engineering, or politics generally
depends on whether you are on the upside or downside of the eventual

>What I was complaining about is that it looks like 'someone' (Slemko?)
>is 'assigned' the duty of being the 'front end' for new messages
>that arrive and more often than not the initial response is inappropriate
>and sometimes even insulting.

Marc has not been 'assigned' anything -- he does that stuff to the extent
that he has time to do so, of his own free will, most likely because he
cares about Apache and if he didn't do it, nobody would be doing it.
My experience, having read several of these cases, is that Marc will respond
in an insulting manner if the person who submitted the report is asking
us to do the impossible -- fix their problem without even telling us
what the problem might be, where to narrow our search, or even what platform
they are using.  Basically, they are wasting our time.  If that means the
user goes off in a huff and buys a competing product, fine -- it just
means they'll be wasting our competitor's time rather than ours.

>I was suggesting that if the 'company' wanted to put a better foot
>forward they should think about 'assigning' this 'first response'
>duty to someone with less attitude and more patience, that's all. 

One of the nice things about Apache is that we don't preclude all the
opportunities for other companies to provide user support.

>>  And if you have not noticed most of the core member really don't care if
>>  microsoft "catches up".  We started this to build a great web server, not
>>  take over the world or make a profit.
>>  Cliff
>I would submit that the core members care very much about 'protecting'
>their 'product'... as they certainly should. Anyone who has 52% market

61% market share (not that I'm counting ;-).

>share of anything should care very much about their 'responsibility' to
>that market. Witness the recent incorporation of Apache to the legal
>entity known as 'ASF' in order to make it easier to 'raise capital' via
>contributions and ( in Roy Fielding's own words ) 'go beat up on
>people who are abusing our name'.
>Whatever the motivations... these are NOT the actions of people who
>take the future of 'the product' lightly.

No, not product -- project.  Products are just the result.  The ASF exists
to protect the projects we work on and the people involved in those projects.
A project is damaged if some other group labels something other than the
work of that project with the same name, in order to benefit from the good
will generated by that project without participating in that project.

>I know the effort it requires to start 'a company' with a legal identity
>and it takes a LOT of work. Most of it is often classfied as 'tedious' 
>or even 'boring' to essentially non-business minded folks such as many
>programmers and academics so it is commendable to see the effort
>being made on behalf of 'the product'. For Brian and Roy and the other
>upper-tier members to even undertake this effort is, I think, an indication 
>of their true committment to the prolonged life of 'the product' in the face
>of increased competition. It is a smart 'business' move for them to make
>at this time and it proves their genuine care and concern for 'the product'.

Damn straight -- a LOT of work.  It was done to protect the project and to
enable other projects of the same type to be formed (eventually).

>As for whether or not upper-tier members care about 'Microsoft'
>( or any other Server manufacturer ) 'catching up'... who can
>really say for sure? The amount of M$, Microsad, etc bashing that
>goes on may be just the usual 'nothing so unites a group as a common
>enemy' sort of commentary that is common to UNIX bred programmers
>and there probably is no point in reading too much into it... but even so

Ummm, just as common among MS-bred programmers.  You don't see it as much
because MS-bred programmers are rarely interested in open projects and
keep their bashing hidden within the hallways.  Microsoft is the best
example of a company that only gets motivated by competition with a
recognizable foe.  Well, actually they are motivated by option values,
but that is directly correlated to their public strategy toward any
potential competitor.

>the most recent posting from Roy Fielding to the members at large
>is worth another look...
>Subj: Re: apr and 2.0
>Date: 6/25/99 3:59:10 PM Central Standard Time
>From: Roy Fielding
>Okay, guys, cut the crap.  The next person who complains about how
>long it has taken, or will take, to get 2.0 out the door had better
>be promising their personal time towards actually working on it
>instead of just talking about it.  Pissing in the wind is not a very
>good strategy.
>Roy Fielding 
>'Strategy'? Interesting word choice. 
>I have seen many a 'shape up or shut up' MEMO come
>down from BOSS to EMPLOYEES and this looks pretty
>classic to me.

If any of these people considered me to be a BOSS, they wouldn't have
listened to what I said.  There are generally two ways to get something
moving: lead it from the front or push it from behind.  Pushing from behind
does not work for a volunteer effort.  Leading from the front only works
some of the time, but that is better than none of the time.  That is what
I meant by strategy, since the only reason these people were complaining
about progress was because they wanted the group to *make* progress, thus
pushing from behind instead of leading from the front.  If Dean, Ryan, and
Manoj want to complain about lack of progress on 2.0, fine, because in doing
so they are identifying areas that are available for others to lead.

I do a lot of things that are more bossy than the rest of the volunteers,
and sometimes I have been rightly criticized for doing so.  But that is just
the way I get things done when no other alternative seems to be working,
or when time is so critical that any degree of self-centered determination
will (hopefully) be understood as only temporary.  Try as I might, that
isn't likely to change anytime soon.  Sometimes, it seems like people are
just waiting for me to say something before they take-off at full speed,
as if my comments were some sort of starter's flag.  Other times people
just tell me to piss-off.  I expect to get both extremes, because I don't
expect to be right all the time.

The Foundation project is a slightly different matter, since I have the
legal responsibility of being the ultimate boss on the administrative
side of things, requiring that I make sure the board acts in accordance
with US+State law and the bylaws.  Fortunately, the first thing I made
sure of is that any decisions that might be interpreted as 'bossy' are
delegated to whatever part of the organization will be affected by that
decision, thus preserving the self-determination of each of our projects.
That is, ultimately, what is important to me.


View raw message