[openbeos] Re: BugZilla => Trac

Hi Axel,

First of all, I want everyone to know that I did not mean to personally attack you or Waldemar. If that is how it came across, I apologize. :-)

Axel Dörfler wrote:
It is a problem. By forcing people to use Trac in the same subdomain as the beta website, you are indirectly legitimizing a beta website that is still in development. You have now created two parallel worlds that are prone to confusion and mixed messages.

Okay, how can anyone take a website under the subdomain "plonetest" for real? Especially when it has been stated publically (in this list) that this is the testbed for the new site?

1) Casual browser clicks on a http://haiku-os.org/bugzilla link (there are many of those links on the web)

2) User is automatically redirected to the Trac page, and invited to "...login with your Haiku website account. If you don't have one, please go to our <url>website</url> and register an account" (sic).

3) User clicks on <url>website</url> link and lands on Haiku website...

...or so the user thinks. In this scenario the user has at worst been misled to believe he/she is at the Haiku website, at the least been confused.

The internet is really a web, and you can't possibly know how many links are out there pointing to one and each of the pages on our website (bugzilla included). Posting something on this list will not prevent the situation that I just described from happening.

to have empowered themselves to make the decision to migrate the website,
I also question that two people (that happen to be engineers) seem
[...]
I am not making attacks. If it came across as a personal attack, I apologize. But my intention is simply to make people aware of the impact that this can have.

That sounds to me like an attack - and it would certainly help me to understand what you think it impacts if you would actually explaining it.
The occasional visitor might have a look at the bug tracker, he might even wonder about that "plonetest" subdomain, and he will also certainly miss a link back to the site (as with the old bug tracker, too). However, that's as far as it goes IMO.

Again, not attacks intended. :-)

It is not a matter of whether you or I think if Trac is a separate thing from the website or not. We know too much about what's going on at Haiku. The mistake would be to assume that any potential visitor knows as much as we do, and therefore would have the de facto understanding that this is a beta site (because it was said on the list, or mentioned in a forum, etc.).

Everyone actually logging into Trac will have a reason for doing so (ie. will have used Haiku at this stage, where it's not really easy to get and try), and only they will see the new website. They probably already know about this, anyway; it's not like we have 200 people reporting bugs and a real userbase.

Sure it has, IMO (and IMO only) it's just not *that* important at this point. It will become much more important in the future.
Marketing may be more important now than after you release Haiku. It is a lot easier to market something that is finished, than something that is in the process of being made (like Haiku now). You may need to make strategic changes in your marketing depending on the development stage you are at, but marketing is always important.

You keep telling this, and I noticed, but why do you think marketing is so important at this stage, when only very few people know about us and care?

Axel, you may not have realized it, but you answered the question yourself! :-)

Marketing is important for Haiku exactly because few people know about it, and because we want more people to know about (and contribute to) the project. It has been said very often that Haiku needs to attract more developers, and I think we all agree on that. If it is well done, Marketing could be a great tool to pursue that goal.

Perhaps you think marketing is only needed when you have a tangible product to sell, and you equate it to producing some CDs, running some ads, and handing out pamphlets. While that may also be marketing, it is only a part of it.

Most importantly, marketing does work with ideas too. Since Haiku does not have a "product" yet, marketing can still work to sell the set of ideas that drive people with the ultimate goal of producing that "product", with the goal of attracting more developers (which is what we want to do at this stage).

If you had already started trying to make Haiku more known in some way, it might be different, but have you?
If you asked me what I have done in my capacity as marcom lead, maybe I should remind you that I have only been on board one week. Please, give me a little bit more time to prove myself. :-)

If you are asking me what have I done to make Haiku known as an individual, the list is quite long, but I do not want to bore you and the rest on the list with what I have done for Haiku.

Frankly, in this situations I feel envy of the programmers. I wish there were CIA bot or RSS feed than could show the many intengible contributions that a lot people make to the advancement of Haiku in many different ways, as there is for code commits. :-)

Also note, that our *current* site is not really ready to attract new people, as it's so outdated. And only very few people have the rights to update it for whatever weird reason. So right now, trying to show Haiku to more people might also just be counterproductive because of the website. And that's why we're a bit in a rush, we left it untouched for too long.

Yes, agreed that the existing website has many shortcomings. I also understand that it took very long to get started. But we let something take a disproportionate amount of time to get started, and then we want to rush it out the door.

I know this may sound too philosophical, but Haiku was inspired in BeOS. BeOS was about attention to detail, and what we are doing that does not reflect that spirit. I have a problem reconciling that feeling with the way we are trying to get the website out the door in a rush.

I am not referring to Trac specifically, but to the new website project as a whole (from several months ago). It leaves a very bad taste and a strong sense of unilateralism when somebody from the outside tries to help on an open mailing list, his opinions are shot down, and then a few privileged decide behind doors to do what this person had proposed in the first place.

This is what happened to me in the haiku-web mailing list. Now, tell me, how is that open? How do you expect people to join and/or contribute if

Would you care to be more specific? I'm certainly not going through the list's archive and try to shut your arguments down. I'm listening on that list, too, and I didn't notice anything in that regard, though.

I never said that you that shut me down. :-) I used the haiku-web list to propose practical solution, one that used a CMS package (I thought about XOOPS, but any other would have been OK). That proposition was met with the stance that all the CMSs sucked, and therefore we would go with Rail Frog. On paper, Rail Frog looked like the utopia CMS that the dreamers had been waiting for, but it was (and still is?) unreleased, and there was no clear indication of when it would be ready.

I warned that it was not smart to put all your eggs in the one basket of an unproven tool; I even proposed that we could take an interim step by using a CMS first, and then if/when Rail Frog was ready any time in the future, a transition could be considered. This fell into deaf ears. I left with disappointment. Two months later, to my surprise, I read that Drupal was going to be used. I almost fell off my chair when I read that!

In a nutshell, I have nothing against Drupal. but the whole process comes across as being capricious and closed.

you are not receptive to ideas from others, and always hide behind the "it's been decided" smoke screen?

Haiku seems to have a very inward centered view of things. Because the admins discussed or voted on something (behind doors), it does not make it legitimate in the eyes of the people looking from outside.

That is the closedness that Haiku has: you setup an open mailing list, implicitly encourage people to contribute, and then make a decision behind doors, no explanations given to anyone outside of the admins. If someone on the list asks "why did you...?", Haiku's typical answer is "it has been decided, so don't beat the dead horse".

Sorry, but I think that's a completely unrealistic POV. Usually, discussions start publicly, and we're listening to their input, and then make a decision. It's neither possible nor desirable to always make public decisions and include everyone - it happens if there is consensus, but that doesn't happen so often. Usually everyone keeps yelling their opinion, but neither party is really listening to the arguments (what? there are even arguments?) if any.

I am not sure that I am making myself understood; looks like I may not be that good of a communicator after all. :-)

How can I put it simple? It is not that I want everyone to take part in the decision process; that is not viable. What I would like to see is some sort of transparency. I believe that the Haiku supporters that spend time in the open mailing lists pouring their opinions deserve it, and that it will also serve to give them a real sense of community.

Perhaps the proposition to publish summaries of the admin meetings could help in this respect. Maybe posting the logs of the IRC meetings is just enough (we may have to change the format of the meetings).

In the particular case of the website backend, it was even me who pushed things, because I was sick of waiting for that dream CMS. And suddenly there was a solution (Drupal) for the problem we had for months. *If* you're uncool with how this went: you had approximately a whole year to make a better suggestion. So where's the rush?

And I am glad that you pushed. As I said, two months ago, I was a proponent of using an existing (and proven) tool, instead of (what I call) the utopia CMS (that I was almost certain would never come).

That is not my idea of openness or collaboration, and something has gotta change. I will propose to the admins that we publish at least a summary of the topics discussed and the decisions made on the weekly admin meetings.

That's fine with me, but someone has got to do it. But I don't really get your point in doing it: what's the difference if you hear from a decision via an admin meeting log, or via an announcement on our website?
I still think that the discussed topics could be of interest to others, as well, though, I just don't see what it has to do with the problem you see.

If we can post announcements on our website, I do not see why we could not post a summary of a meeting; if it is an IRC log, that would even be easier. It has to do with the transparency of the organization.

All your points about the existing website are taken, but since you are redoing it, try to do it right. Last week, I think I pointed out the problems of communications that Haiku has on the mailing list, and this is a great opportunity to address those issues. Please, give it chance to happen.

I'm all in favour of doing it right, but since Waldemar is trying to push the new website, we have to talk about this, too. Why exactly don't you think it's not ready yet? I'm sure Waldemar is open to arguments. Discussions can only work right if you make constructive arguments.

I will put this in a separate email.

I did not say Waldemar was sloppy; I said that his rushing of the transition makes the project look sloppy. You seem to take it lightly that he decided to make the switch to Trac without announcing or even consulting it with anyone else (that's the unilateralism and lack of collaboration), but I don't, because I think it does make us look sloppy and disorganized.

It makes us look sloppy and disorganized where we actually are disorganized and sloppy, obviously. I agree with you that Waldemar could have done it better, but it's only the bug tracker; he didn't accidently release Haiku or married Michael Phipps without asking us.

Again, obviously you view these things more lightly than I do, perhaps because you are not in marketing (nothing wrong with that).

Anyway, what's the rush to migrate today no matter what? Why can't we wait a few more weeks until we get it right? We have waited 5 years for Haiku, and we cannot wait a little longer to get a website right?
We can, yes, the question is just (apart from the looks): why?
For the same reasons that you would not release Haiku 1.0 now and say "well, it is better than it was last year; we will improve it little by little" and smile away. It is simply not ready.

Sure, but consider this: it's much better than the previous release. That one continues to give other people bad impressions; if the new (and unfinished) website makes a little less bad impressions, that's already an improvement.
So if we'd make the transition now, we would need to make it clear that this is not the final site yet, but that we'd felt that a visit to it is already more worthwhile than one to our current site.
And Haiku today is much better than a year ago, but we still do not release it, and for good reasons. I apply the same good reasons to the website.

Again, perhaps because you are not in marketing, you view these things more lightly, and therefore the disagreement.

I am going to finish this email the same way that I started it, by saying that I had no intentions of offending anyone, and if I did, I apologize. Hope we are still all good friends. :-)

Koki


Other related posts: