To this list:
I've been advised that in a situation like this it isn't wise to keep emails off the list, so I'm bringing a private conversation back into the list, since it has gone a lot further than I had expected.
Stas Bekman wrote:
This is just a software. This is not real life. So making enemies based on software disagreement is at least foolish.
Not real life? I have a company that is currently trying to convince customers that mod_perl2 is going to be a good enterprise platform, and that perl and thus my company build the best web applications.
But mp2 doesn't have a problem. If your customers choose mp2 it'll just work. The whole issue is about edge cases, which I fail to see how they possibly affect your business.
But that is demonstratably wrong. Not having access to any of the perl standard perl infrastructure is a problem. I have to tell them.
I've built this system, but if you ever need to fix something, don't use any of the standard tools because mod_perl2 is incompatible.
I don't know about you, but I consider my career and my business real life.
I didn't mean that. I meant the API choice. If you tell me that the API naming will mean death or life for your business, I'm sorry, I can't offer you much help here.
It's hardly life and death, and the API choice isn't the problem. If you could make the API choice without requiring so much incompatibility and damage, this wouldn't be a problem.
If you don't accept that a problem is even a problem, and it appears to me from your responses that you don't understand why it is a problem, then I shouldn't try to convince you?
What's wrong with having a different opinion from yours? As it's seen from my tower there is no problem. There are a few inconveniences, but a problem?
Then maybe you need to come down from your tower for a little while. These are not small inconveniences, these are big inconveniences. And while YOU may think them small, I assure you that other people do not.
And these same people can suggest solutions to the problem to you, and are willing to help, and you refuse to even wait to let them help.
Now again, don't get offended, but you need to wear my shoes to be able to judge my position. I know what's the best for modperl users, even though it might be against your position. And that's fine. You have to admit that you are in no position to judge our decisions, since I don't remember you contributing a single bit to the modperl project.
Well, for starters nobody can work on _every_ project. And every time I've asked people how mod_perl2 was going, I had been assured it was being taken care of...
How is that our fault? The current scheme was out there for 3+ years! The fact that you've relied on someone else to do the assurance, is not our problem. You are too late into the game. Release Candidates are out, people already run their businesses using the current API, since RCs indicate no API changes (well as little as possible). Why do you allow yourself to make a decision on behalf of these users, and force them to rewrite their code, because you aren't happy about it? Have you considered that?
Yes I have. I've talked to some of these people who have indicated that they don't mind.
These early adopters go into this knowing things might change, and they represent around 15% of your total user base. You owe your ENTIRE userbase to produce a better designed and better integrated system more than you owe the early adopters there will be no change.
In any case, I'm not suggesting wholesale changes. I don't suggest any changes in the actual API itself or the underlying implementation at all, just that you don't clash with the old one.
For the most part, it would just involve s/Apache::/Apache2::/
And until a production release is out, people understand that there might be change. While it may be my fault for waiting as long as I have, that is no excuse for not dealing with problems once they have been presented.
On the other hand I've spent 6 years of my life, doing modperl day and night. I'm not offering pressing charges against your projects. And I think you should stop doing that for ours.
It's only "pressing" because it would appear that I'm not being given enough time to do some better analysis and provide some proper solutions. I'm seeing huge problems and being thrown into a situation at extremely short notice to try and provide solutions to problems I'm currently absolutely confident I can fix, while working against the clock from a laptop in my care while I'm supposedly on holidays.
And some of these problems can't be fixed once you release properly.
As explained above, unfortunately, you've missed the train. If you'd have come up earlier everything may would have been different.
Until you have done a production release you can make any decision you like.
You need to earn a voice in a community where you try to enforce changes. You can certainly offer things, and we listen to suggestions just like from any other user. But you must not demand things if your suggestions aren't accepted.
All I demand is time.
Sorry, but we have no time. Let me explain something. modperl is being developed on almost 0 resources. Most of developers are very busy and have no time to develop or support the product. I was the lucky one to have windows of full time work on modperl, but in 2 weeks I go back to school and there will be almost 0 resources to push for a release, deal with bugs, questions and writing docs. Now you understand why I can't wait. It has been dragging for too long and it must come out now. Any other delay is playing against us. Many users have expressed this as a problem, since they can't even start moving to mp2 before an official 2.0.0 is released.
As you run a business you know what a deadline is. We had a deadline set on dec 2004, we have missed the deadline already. So if someone did something wrong, it'll be fixed in the next iteration. But the product must ship now.
How many times must I say that this problem is something you CANNOT fix in the next iteration. The two core problems are something you are going to have to live with indefinitely.
I'm afraid that I can't stas. But I'd rather work with you than against you.
Me too, but the circumstances aren't right to do any change now.
People have already volunteered to help with any changes.
BTW, whereabouts in AU are you?
I live in a place called Lismore, in northern NSW. I'm currently in Newcastle in a Internet cafe with zero access to my normal development gear with which to provide you actual solutions.
I haven't been to NSW, I've spent 6 month in Melbourne 1.5 years ago and I loved it. I'm planning to come back one day.
Again, give me the time to deal with this properly, and I'd be more than happy to back off from my current position.
I'm not an unreasonable person.
I'm not saying you are, but I hope now you understand my attitude.
I do. It's obvious that you feel you are irrevocably committed to this release, and that nobody else has the chance to change this.
On one hand I'm being told that it's the mod_perl community that controls this stuff, and on the other I'm being told by you something different.
If others in the mod_perl community want to at least pause to check whether this can be solved, what am I supposed to believe...
Indeed, I'm probably more reasonable than most of the people that have been talking to you until now, like Randal.
:)
But I _hate_ working against the clock.
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
