rc2 and 1.0

Andrew · 2005-01-18 04:37:45

rickg wrote: “_And another thing â

Last edited by compooter (2005-01-18 04:45:07)

rickg · 2005-01-18 04:58:15

compooter,

Sure, and you can check it out from source control, but you cannot download it as a release. And if you’re about to tell me how easy it is to check out then why hasn’t someone done that and packaged it up as an official RC2? I’ve done software dev for 15 years – RC2 refers to a specific build not “everything since build 40.” And, quite frankly, it’s apparent RC was declared prematurely – a release candidate by definition is something that’s feature complete with no known A, B or C bugs – a release that’s a candidate to be, in this case, 1.0. Not something that’s had 20+ builds since RC1.

Dean – if you read this, please don’t take it as criticism. I think TXP is wonderful software. But PLEASE… communicate out what’s going on once in a while and let us know what’s coming up.

To Destry’s point – I actually don’t care that much either – but after seeing B2 go this way it makes me nervous to see still another development effort not talk to it’s community. And I want TXP to move forward and stay relevant, not fall behind or lose people because they are not sure what’s coming up.

duffe · 2005-01-18 05:28:46

Not all of us are getting it for free. Some of us are supporting its development through Textdrive hosting. Others are contributing plugins and documentation that increase the value of TextPattern as a whole.

Now don’t get me wrong, Textpattern is brilliant, but the development process is terrible. It reflects badly that there is no stable release available for download on the homepage. It reflects badly that a release candiate in September has not progressed to a major version by mid January because, presumably, new features (themes engine?) are being added. It reflects badly that ETAs are continuosly not met.

Furthermore, is 1.0 actually going to be beta-tested by the community? Sure, RC1 & RC2 have been stress tested, but what about the new parts added to the codebase since. I’m worried that 1.0 may not be the stable release I’ve been long-waiting for…

I know there is no warranty on GPL’d software, but things here really aren’t good enough if Textpattern is to be considered “serious” software.

Sorry that my first post here is a bitchfest.

Method · 2005-01-18 05:45:00

rickg, you’re right to ask questions, and wilshire, you’re wrong to dismiss the need to ask those questions. Patience is always a good recommendation, but there’s really no sense in the way that Dean keeps a closed mouth, doesn’t invite collaboration, and generally tempts people to ask the potentially deadly question, “is this project dead?”

Here’s why: Dean’s slice on the open source business model is to provide hosting services and technical support based on textpattern. He also solicits donations from the front page. Labor of love or not, Dean is also “in it” for the money. In fact, that’s why Dean has maintained such tight control over the project so far, because he wants to ensure that textpattern remains a viable product for him to build his services on. Not a hint of criticism yet, because what I’m trying to point out is that free software is, or can be, a business. The success of that business depends in large part on its success at increasing the number of people using the piece of free software, and in making sure that those people are both generally satisfied with the product and confident in its future progress. Because, again, Dean makes money when people think that textpattern is so good that they want technical support from people familiar with the software’s inner workings, or when the quality of the total software “experience” is such that it creates enough goodwill to make someone want to make a donation or sign up for a hosting account.

Bad communication is bad business. Because of a lack of information, by which I mean reassuring words and displays of project leadership in addition to concrete timelines and feature promises, my confidence (read: faith) in textpattern has fallen, so that I have found myself revisiting Wordpress et al. (Wordpress, in particular, has great public relations). Now, one could respond “Texpattern is free! Youâ

reid · 2005-01-18 05:46:48

> rickg wrote:

> RC1 is SEPTEMBER? And no RC2 for four months? Does this mean txp is not a priority?

How about a little perspective? From MT 2.6x to MT 3.0 was over a year. From MT 3.0 to 3.1 was nearly six months. Right now, there’s absolutely no feature list available for MT 4.0. From a corporation with a staff rapidily approaching triple digits.

From my viewpoint, I’ve been using Textpattern 8.5 months, and have seen four versions during that time.

Jeremie · 2005-01-18 06:08:04

> duffe wrote:

> Not all of us are getting it for free. Some of us are supporting its development through Textdrive hosting. Others are contributing plugins and documentation that increase the value of TextPattern as a whole.

That don’t give us a “right to demand”.

But, beside the money aspect, I think the most important contribution is code. There are several real good developpers around TXP, but yet they don’t seems to be able to inject code, propose solutions, and so on. They have to freeze existing or future plugins dvlp because they don’t know what is going to be fixed, change, and so on.

I am certain that in a snap of a finger Dean can get at least 2 or 3 real (read professional) developpers who would be happy to contribute with code.

One example ? Clean URL. They are hundred of people having issues with it, and just on the forums we have seen dozens of better ways of handling it for months. So discuss it, implement it, increase the version number, release and be done with it.

And so on, and so on, and so on.

>It reflects badly that there is no stable release available for download on the homepage.

imho it reflects badly that it’s called a release candidate and not an alpha version, or an early beta one. Developement take times, if it’s not a production product yet for the purpose intended by the author it’s not a shame not to have a 1.0 for download.

> It reflects badly that a release candiate in September has not progressed to a major version by mid January because, presumably, new features (themes engine?) are being added.

I’m not sure about a major one, but no new version at all (well half one, not publicaly disclosed and requiring tech savvy skills to handle), yes.

> It reflects badly that ETAs are continuosly not met.

Well not per se, shit happens to all of us, but there are so little relase and ETA … :-)

> Furthermore, is 1.0 actually going to be beta-tested by the community?

That is the purpose of a Release Candidate in theory…

> Sure, RC1 & RC2 have been stress tested

Bug reported, but no fixes published. So it’s impossible to tell if the fixes are working for a large scale base.

> but what about the new parts added to the codebase since. I’m worried that 1.0 may not be the stable release I’ve been long-waiting for…

I’ve found TXP pretty stable so far. I’m not worried about stability, I’m worried about conceptual problems, and some bugs. And some lack of features.

Yes I know, features are not everything. So let’s talk one simple example… take a non-xhtml-guru (like you ‘ma, or grand’ma for the youngest) and put it in front of TXP. Ask her to write an article like she would for a newspaper. If she could, the result won’t be semantic and clean. Why ?

if she’s not able to read english, she won’t get much help from Textile and TXP help since it’s not translatable.
she can’ write proper introduction, inserts, post-scriptum (and the like) because there are no ways of explaining on the backend what class or id she has to use (or if you prefer because there are no way of assigning class or id to custom field output, because the size of the field entry is very small, hidden and not in the flow of the other writing area)
she won’t know what categories to use in a heavy and complex web site, or more probably she will use the wrong ones, because there are no meta data on section and categories, and no drop-down fixed choice for some custom fields.
you can’t leave her advice, tutorial or guides because there is no way of communication in the backend. You could use a external forum, or mailing list, but there are no ways of integrating these â

Last edited by Jeremie (2005-01-18 06:14:02)

wilshire · 2005-01-18 06:12:23

I’m not disagreeing that knowing what’s going on would be nice. And you don’t have to tell me about all the people that have contributed plugins and documentation. Its just that the impression that I’ve gotten from some of these posts is that Textpattern is a life and death matter. NEWSFLASH! Its not.

I’m looking forward to 2005: the year of the text, just like the rest of you. I guess I and others like reid are in the minority because we’re thankful for the progress that’s been made thusfar and willing to wait for annoucements to be posted instead of begging for them.

And for some more perspective, head on over to sourceforge and see how many projects were posted and then never updated again.

rickg · 2005-01-18 06:26:38

reid,

Please don’t get me wrong, I like txp. But your example is of released software, not time between pre-release builds. Using your context, I don’t care if, after 1.0 it takes 6-12 months until we see 1.x or 2.0.

But, as I stated above, RC1 is an indication that you’re a few builds from the 1.0 GM milestone – not 25+ builds and over 4 months away.

Look, what I’m asking for is

1) a real date for 1.0. Sure, I’d like this tomorrow, but I really don’t care if it’s a month from now. But presumably Dean’s got a list of what he feels needs to be done before 1.0. Is that a month worth of tasks? A week? A day?

2) a roadmap for what’s going to happen over the next, say 12 months. Put it on a wiki and make sure everyone understands that this is an evolving process. When Dean makes a decision to change something, he edits the wiki.

3) after the release enters beta, a post every week or 2 updating the community on what’s happening (for example, “ we should be seeing Beta4 in a couple of weeks…”

4) weekly or monthly builds (archives, like RC1 on the front page) as appropriate. Don’t post weekly builds if nothing’s really changed, but let’s not go more than a month without a new build once we’re in beta.

rickg · 2005-01-18 06:35:17

Wilshire,

Of course txp is not a life and death matter. Never said, or implied that it was. But if Dean wants it to catch on he’s gotta talk to us. BTW, I’m not begging for anything… and I have been waiting, just like you and reid – for weeks and weeks. But at some point I’ve gotta decide if this is going to be an ongoing effort that I should use on client sites or if it’s going to be like B2… a cool product that was mysteriously abandoned by the sole developer.

Last edited by rickg (2005-01-18 06:36:05)

Damelon · 2005-01-18 07:37:52

I believe Dean has said that development would be opened up to additional developers after the release of 1.0 (I’ll take back this statement if I am misremembering). Until that point question, complain, bitch, whatever, but Dean isn’t really misrepresenting anything.

As for his release date: I hate that developers do this. They say a date, they believe it’s accurate, it virtually never is. Dean is far from the only guilty party in this area.

As for making lists of things you want (expect) Dean to do…

rickg · 2005-01-18 07:50:48

Damelon,

What, a list like mine is rude? Is that what you’re saying?? Look, don’t put up a community site so that people can talk to you, then complain that not everything we post is laudatory.

And, what, precisely, is unreasonable about the list I put up? It’s simply my opinion, what I think Dean should do based on a decade and a half of managing software projects. He can ignore it, laugh at it, print it out and shred it, whatever. Sorry, but given the almost total lack of outbound communication, I’m not sure anyone should get too uppity about someone proposing a few ideas.

Damelon · 2005-01-18 08:07:49

The beautiful thing about the ellipsis is everyone will make of it as they are disposed. But since you have shown yourself to have a short fuse I’ll elaborate.

Dean knows all about work-flow, scheduling, project management, communication, and all the rest. You may not feel he handles it in a manner consistent with your views. Fine. But the manner in which Dean works there is a consistency. We go through cycles around here and until Dean adds people to the mix that cycle is his.

Don’t bother arguing the point with me, I may have a certain sympathy for your viewpoint, but I think you expect too much.

rickg · 2005-01-18 08:14:55

But then how can we know what’s going on? Look, I really don’t have that short of a fuse, and I really DO like TXP, but I simply don’t understand why having a periodic update is that hard.

Anyway, back to waiting…. ;)

koopd · 2005-01-18 10:54:46

Any chance of a little update from the man himself on the matter?

duffe · 2005-01-18 12:21:51

Jeremie wrote:

> That don’t give us a “right to demand”.

Well it does to a degree because there are some levels of service that need to be adhered to in order to keep customers on board. Look what happened to Moveable Type when they kept customers in the dark about license changes last year. A fair chunk of their user base moved on to WordPress as a result. With that in mind, there are some “demands” that Dean needs to satisfy his customers with in order to prevent losing them.

> But, beside the money aspect, I think the most important contribution is code. There are several real good developpers around TXP, but yet they don’t seems to be able to inject code, propose solutions, and so on. They have to freeze existing or future plugins dvlp because they don’t know what is going to be fixed, change, and so on.

Absolutely, that’s the source of my frustration. I’d like to improve the search engine for a client of mine and release the changes back to the community, but I’m reluctant to do so if Dean has made any improvements himself. If he hasn’t then I still have to port the changes across to the new version, and I’d rather not have to that given that a new version could be released at any moment .

>> It reflects badly that ETAs are continuosly not met.

> Well not per se, shit happens to all of us, but there are so little relase and ETA … :-)

Dean has got our hopes up quite a bit about an eminent 1.0 release and has thus far failed to deliver (whatever happened to the nightly tarballs?) Sure, shit happens — but a track record of shit happening is not good for business.

>> Furthermore, is 1.0 actually going to be beta-tested by the community?

> That is the purpose of a Release Candidate in theory…

I know, that is my point. Supposing that new features have been added since RC2, will a RC3 be released to accomodate testing of new features, and will we have to wait another x months before the bug necassary fixes are implemented? If thats the way Dean wants to run things then great! But please document such schedules so I can make good decisions about using this software.

>> I’m worried that 1.0 may not be the stable release I’ve been long-waiting for…

> I’ve found TXP pretty stable so far. I’m not worried about stability, I’m worried about conceptual problems, and some bugs. And some lack of features.

Sorry, I used the wrong term. I meant to say rock-solid: as in barely any bugs. TXP has been superb so far as pre-release software — but I worry about a 1.0 release tested only by the developer..

> I know there is no warranty on GPL’d software, but things here really aren’t good enough if Textpattern is to be considered “serious” software.

> I’m not sure that’s motivation for Dean. He pays his rent with Textdrive, not Textpattern (and by the way I would not mind pay some reasonnable fees for it, yes “free” open source software allows it, and I’m not the only one.. Zem’s ransoms have proved that), and if Textdrive is not very succesfull I’m sure paying the rent has priority over TXP (and I’m guessing that’s what happen in 2004).

I think you are spot on here, unfortunatly. I’m hoping TXP is indeed opened-up to more developers post 1.0.

Textpattern CMS

Textpattern CMS support forum

#91 2005-01-18 04:37:45

Re: rc2 and 1.0

#92 2005-01-18 04:58:15

Re: rc2 and 1.0

#93 2005-01-18 05:28:46

Re: rc2 and 1.0

#94 2005-01-18 05:45:00

Re: rc2 and 1.0

#95 2005-01-18 05:46:48

Re: rc2 and 1.0

#96 2005-01-18 06:08:04

Re: rc2 and 1.0

#97 2005-01-18 06:12:23

Re: rc2 and 1.0

#98 2005-01-18 06:26:38

Re: rc2 and 1.0

#99 2005-01-18 06:35:17

Re: rc2 and 1.0

#100 2005-01-18 07:37:52

Re: rc2 and 1.0

#101 2005-01-18 07:50:48

Re: rc2 and 1.0

#102 2005-01-18 08:07:49

Re: rc2 and 1.0

#103 2005-01-18 08:14:55

Re: rc2 and 1.0

#104 2005-01-18 10:54:46

Re: rc2 and 1.0

#105 2005-01-18 12:21:51

Re: rc2 and 1.0

Board footer