RE: Unfulfilled promises... forever experimental extensions... all over again

php.internals

NAIK,ROSHAN (HP-Cupertino,ex1)

22 years ago
Rasmus wrote:
>> Really ? Are you sure ? >> http://www.zend.com/zend/week/week146.php
> Ok, one self-serving fix.
I was (again) expecting exactly that kind of response. I dont understand why so much arrogance exists among a select few. And it unfortunately reflects the community's attitude. So nobody till date or in future will benefit from the fix right ? A self-serving fix may be today. But a problem waiting to happen on other platforms anytime(if not happened already). But lets leave that issue off to the side, its a red herring as far as the central issue here is concerned. Rasmus wrote:
> So, how about proposing an actual solution instead of pissing > off the very > people who you will need help from to improve things?
I dont beleive all the people are being pissed off. They are just not speaking up their concerns (given up hope?) I dont see why anyone pointing out a problem is pissing _you_ off. Perhaps then every individual filing a bug report should be pissing you off too. They should be asked to go fix it themselves...and not bother anyone here as they are not part of the "inner-circle" or "the community". Rasmus wrote:
> Some don't contribute anything, but they > are still part of this community. By being part of the community you have > a voice and through our extremely open approach you can easily contribute > your code and ideas.
In theory. But clear evidence to the contrary here. Rasmus wrote:
> This > solution may > be a set of general criteria an extension has to meet to leave its > experimental state, a set of test tools, or a group of > developers that you > convince with your obvious social skills to contribute to this effort.
I am glad to see a slight progress from "please bring up concrete problems" to "please bring up actual solutions" I already proposed a couple solutions in my 1st and 2nd emails. Please review them again. More solutions will come once people take up the problem seriously. Thomas wrote:
> Actually, MS has a responsibility to make money, and as you said, there is > no way get everybody over to FF.
Also it is not entirely correct to say that no one behind the curtains of PHP internals isn't making money off of PHP. Some do some dont..directly or indirectly. But this topic is also a red herring to the central discussion. So lets leave it alone. Rasmus wrote:
> Then you better put on some protection, because this isn't going to > change. Like I said, we develop software for ourselves, not for > customers.
Not all opensource projects suffer from the same problems that PHP (more accurately a select few here) is exhibiting. As the previosly quoted user said .... <quote> "Open Source" is a philosophy. It shouldn't be an excuse. </quote>
> We welcome contributions, but we react rather badly when > people come along and demand that we volunteer more of our time.
So far no one demanded to volunteer more time. Just requested to prioritize things that more important, higher. In much of the propaganda about PHP5 I read, Did i read anything outright calling out the fact that "php5 is not backward compatible" ? no! Nor did I see anything that said, "hey we have these new features, but you have to fix them before using them". If someone else in "real life" did that to you... you would be upset too. PHP releases seem to be driven more by whats important for the "oomph factor" to make a big splash. Clearly something has to be learnt from the other lowly languages. In another mail (about apache2 support) I saw PHP's virtues being extolled as "PHP is the glue". And that its strong point is that it allows users to tie all kinds of disparate functionality ( xml , image editing, etc ) together. The quality of most of that "glue" and its usefulness is being question. And some evidently dont like it. Andi wrote:
> I suggest we stop this thread now. It's just taking up time we could use to > more fruitfully in improving PHP. Please move any further rants off the > internals mailing list and keep them personal.
So these are all "useless rants" and wont improve PHP ? I hope its a personal opinion. This is a nice way to snub off concrete problems. Andi wrote:
> I do think that if there are some concrete suggestions to be heard that's > fine, but try and keep it short, to the point, and constructive.
So none of the suggestions were constructive ? or perhaps you didnt read them. Andi wrote:
> Personally, I agree that the "Fix it yourself" argument is not a good one > and isn't always relevant
Acknowledged several times before...but agressively misused till date. Andi wrote:
>, but if you'd look not at what developers say > after you piss them off with such emails,
Why does it piss them off ? What is "such emails" ? This is nothing personal. A select few are indeed taking it personally. But I dont beleive this reflects the community at large. This is purely bringing up an issue that is a problem in "real life" and is not a problem for "the experimenting engineer". Andi wrote:
> but how the PHP development team > has worked in real life,
Thats a blanket generalization...to the point of aggressive ignorance and silly complacency. This is nothing but a "real life" problem we are talking about. So far.. some people seem to be interested in attacking peripheral issues like ... "you didnt contribute a damn thing" ...."this is a useless rant" "nobody is making money here" ..."fix it youself" ..."not a good quote" ... "thats a self-serving fix" In another 2 years (on the 6th epxerimental birthday for sockets ) and PHP6 is released, everyone will have forgotten about what PHP5 didnt deliver ...or even less of what PHP4 promised & didnt deliver. Somebody will again notice the repetition of promises and point it out ... and get snubbed. Such snubbing off is the prime reason why most people prefer to stay quiet right now. I remember aggressive (but weak) justifications to why PHP's OO model was the right way for PHP .. and now here we are! Hiding (in the bowels of documentation and source code) the fact that the "banner features" are not ready ... is dishonesty. Lets not do that. -Roshan

Andi Gutmans

22 years ago
Oh boy, if you don't see the difference between "concrete suggestions to be heard that's fine, but try and keep it short, to the point, and constructive" and the email you sent then it is really preferable to everyone here for you to stop emailing this list. Andi

Zeev Suraski

22 years ago
Roshan, I have to say I agree with many of the points you're making, and disagreeing with quite a few as well. Since I do agree with Andi that for whatever reason, this thread has turned non-constructive, I won't reply point by point. The main point I agree with you about is that it doesn't make sense for extensions to stay tagged with 'experimental' status for years and years, as long as experimental means what it means today. And today it means at least one of two things (a) unstable, and (b) API subject to change. The key problem, the way I see it, is with (b). It's not realistic to tell people that they cannot rely on APIs that have been available for two years because nobody bothered to update the extension status, or because it was 'dumped'. IMHO, we need to add a few more statuses. I believe that most of the experimental extensions fall under only one of the above categories, and we should make it clearer. Doing something along the lines of giving each extension some sort of stability rating, maybe based on user feedback and maybe just based on the gut feeling of the author, and separating the 'API subject to change' from it, would make good sense. I strongly and strictly disagree with you as to what the developers should spend their time on. I'm not sure what the person who said 'Opensource is a philosophy, it's not an excuse' had in mind when he said it, but there's one thing you cannot do when you get something for free - and that is demand. Whatever someone chooses to give you is solely at their discretion. And if someone feels great spending his time working on optimizer 0.000001% of performance in soundex() instead of fixing bugs in mysql_connect(), the only things you can do about it are (a) email internals@ politely, hoping to get someone to do you a favour and fix it, or (b) find someone that will do it for you, for a fee. At 00:14 27/08/2004, Naik, Roshan wrote:
>[snip] >Hiding (in the bowels of documentation and source code) >the fact that the "banner features" are not ready > ... is dishonesty. Lets not do that.
None of the 'banner features' of PHP 5 is in an unusable state at this point. There's no dishonesty there. At best, if some are still tagged experimental, it means our system is not that good yet. Zeev

John Coggeshall

22 years ago
On Thu, 2004-08-26 at 17:14, Naik, Roshan wrote:
> I dont beleive all the people are being pissed off. They are just
I didn't even bother reading past this line, and I'll be surprised if anyone else responds to this because chances are you've already been filtered out of existence. You're demanding, confrontational, make no attempt to understand the environment under which you are trying to get something done and worse of all annoying the hell out of us all. So yes, it's pissing me off too. You walked through the door trying to fix one thing -- you'd like to see some extensions in PHP get their experimental status removed. The reality of the situation is that they're that way for a reason most of the time and if you don't like that -- well, there isn't a damn thing you or I can really do about it besides change it ourselves. The author of the extensions are the ones who really are responsible for those changes so you should be contacting them directly with your complaints anyway. No one is really "happy" with the idea that some extensions are in this state, but most of us have a full-time job we attend to during the day and then hack on making PHP better when we have the time to do so. If you don't like those rules go use ASP -- I'm sure Microsoft has a great support contract for a few thousand dollars a month where they will wait on your demands. You aren't going to get anywhere with this so just drop it. There are no promises that something will or will not happen in open source, there isn't any promise that anything even works. The bottom line here is this is *YOUR* problem, not mine or Rasmus' or anyone else's here on this list and you need to start treating it that way. You could have gotten a lot farther just asking what you could do to help get those extensions out of Experimental and doing it. Chances are if you would have gotten some help along the way too. Learn some interpersonal skills -- especially with people who owe you nothing.