SMF release schedule??

Started by Daktyl198, October 17, 2012, 10:22:46 PM

Previous topic - Next topic

Daktyl198

Now, I know there's no such thing as a release schedule for stuff like this but...

Its come to my attention that in the beginning of next year it will be the 10th anniversary of SMF. And SMF has yet to reach 2.1-alpha2.

I am in no way trying to criticize the way things are done: One of the reasons I use SMF is because of it's stability and security which I doubt could come from rushed code.

I just feel that it's a little (and by little I mean medium/highly) behind in several ways. There is a good chance it's using some very outdated code and, not to mention, a lot of the themes (especially the default one) are a bit lacking for modern looks.

I've sampled the latest from the GIT repository of the 2.1 alpha release branch and it's looking to be amazing: but I have the same concerns. Since it's been in development for a while, it my also be using some legacy code.

And then when will SMF 3 come out after that? another 6 years? When PHP has either advanced far beyond what it is now to keep up with the hardware or dropped out of favour?

Once again, I am not trying to criticize, I'm just looking for answers.

P.S. It irks me that whenever my friend and I debate over why I use SMF and he uses MyBB, all my points are mute because most of the stuff that mybb has now, SMF is getting in 2.1. And the MyBB 1.8/2.0 releases are looking to be outstanding. I end up having to fall back to security as my main argument for using SMF.

Arantor

QuoteNow, I know there's no such thing as a release schedule for stuff like this but...

Before I say anything else, that's going to be your answer in a nutshell: there's no release schedule.

QuoteIts come to my attention that in the beginning of next year it will be the 10th anniversary of SMF. And SMF has yet to reach 2.1-alpha2.

So what? In that ten years, it's seen multiple churns of development teams, and 5 years of that was 2.0 on its own. That's the joy of having perspective.

QuoteThere is a good chance it's using some very outdated code

So what? There are companies quite happily running decades old code because it works, it's been proven reliable and if it isn't broken, why fix it?

Quotea lot of the themes (especially the default one) are a bit lacking for modern looks.

That's a matter of perspective.

QuoteSince it's been in development for a while, it my also be using some legacy code.

Why would you rewrite code that's been proven mature and stable just because you don't want any 'legacy' in there? You'd be among the first to complain if it were unstable and buggy, I'm sure.

QuoteAnd then when will SMF 3 come out after that? another 6 years? When PHP has either advanced far beyond what it is now to keep up with the hardware or dropped out of favour?

Considering that SMF 3 is being built exactly in line with how you're thinking - i.e. completely clean rebuild from the ground up, how long do you expect it to take to iron out all the kinks, shake down and make sure it is stable, mature and proven? You can't have it both ways, but you seem to want to.

QuoteOnce again, I am not trying to criticize, I'm just looking for answers.

Seems to me that you're not looking for answers, you're trying to have your cake and eat it. You can't completely churn out legacy code every generation, because you'd never battle harden everything. See also vBulletin - 4.x is a complete rebuild and buggy as sin, 5.x is going EXACTLY the same way. Would you rather have mature, proven and stable code or that sort of mess?

QuoteP.S. It irks me that whenever my friend and I debate over why I use SMF and he uses MyBB, all my points are mute because most of the stuff that mybb has now, SMF is getting in 2.1

And MyBB is still a confusing mess to operate. Funny how, if you notice, SMF users don't preach to MyBB about how good SMF is, they generally accept that others have their own opinions, but MyBB users sneer down at anyone who isn't using MyBB. That alone, for me, is enough reason not to move.

Mind you, if you want to take the security angle, I found it hilarious when the 1.6.4 bug came along because the exploit wasn't merely an exploit of a vulnerability, it was an exploit of insecurity by design, and I saw how quickly everyone tried to argue that it wasn't.


I would note, there is an SMF fork out there called Wedge. In over two years that hasn't gotten to an alpha either, despite doing a lot of overhauling of code and pruning legacy stuff and there are still tens of thousands of lines of code barely touched in that time.

Antechinus

Quote from: Daktyl198 on October 17, 2012, 10:22:46 PM
Now, I know there's no such thing as a release schedule for stuff like this but...

Its come to my attention that in the beginning of next year it will be the 10th anniversary of SMF. And SMF has yet to reach 2.1-alpha2.

Probably wont bother either. The current feeling is to just go straight to beta. Not much point calling something Alpha 2. It's either alpha or it aint, if you know what I mean. TBH the alpha is pretty much as good as the 2.0 betas anyway, in terms of stability.


QuoteI just feel that it's a little (and by little I mean medium/highly) behind in several ways. There is a good chance it's using some very outdated code and, not to mention, a lot of the themes (especially the default one) are a bit lacking for modern looks.

Some legacy code is being ditched. Quite a lot of it, in fact. All the stuff for old browsers is a case in point. Other bits of the existing codebase (do you define existing codebase as legacy code?) are being overhauled. Themes are up to themers. There's not a lot that SMF itself can do about that. :)


QuoteI've sampled the latest from the GIT repository of the 2.1 alpha release branch and it's looking to be amazing: but I have the same concerns. Since it's been in development for a while, it my also be using some legacy code.

Not really. 2.1 hasn't been in dev all that long. You can't expect the whole codebase to be written last week, and even if it was it would still need a lot of testing before release.


QuoteAnd then when will SMF 3 come out after that? another 6 years? When PHP has either advanced far beyond what it is now to keep up with the hardware or dropped out of favour?

I have made an educated guess about when 3.0 is likely to come out, but I aint making it publicly. 2.1 will be using up to date PHP as much as possible (bearing in mind that hosts do not always update immediately). Minimum versions have been raised over 2.0.x, for example, but the finished product still has to be suitable for a wide range of users on a wide range of hosts.

Daktyl198

Looking over my original post it ended up sounding very... whiny...

QuoteBefore I say anything else, that's going to be your answer in a nutshell: there's no release schedule.
Yeah... I figured that much at least. I'm used to it.

QuoteSo what? In that ten years, it's seen multiple churns of development teams, and 5 years of that was 2.0 on its own. That's the joy of having perspective.
And I am very glad for the perspective of so many. I love it when more and more people get together (though maybe not all at the same time) to work on something.

QuoteSo what? There are companies quite happily running decades old code because it works, it's been proven reliable and if it isn't broken, why fix it?
Because it WILL become broken. It's just a matter of time. As new versions of PHP come out that people will need for newer versions of other software, the code in SMF may become unusable on the same server. THAT is my main concern that I meant to point out through my post :/

Quotea lot of the themes (especially the default one) are a bit lacking for modern looks.


That's a matter of perspective.
Granted, I have found more than one that I like very much (usually from Crip). But in between those I found many that just changed a few pictures from the default theme. It left me wondering about how many customizations the current SMF allows for themes.

Quote
Why would you rewrite code that's been proven mature and stable just because you don't want any 'legacy' in there? You'd be among the first to complain if it were unstable and buggy, I'm sure.
Once again, I didn't mean to sound like a condescending... uh... *bad word here* about it. It's the same point as above: just because it's not broken now means nothing. Given a length of time, at least part of it is bound to become broken in some way.

Quote
Considering that SMF 3 is being built exactly in line with how you're thinking - i.e. completely clean rebuild from the ground up, how long do you expect it to take to iron out all the kinks, shake down and make sure it is stable, mature and proven? You can't have it both ways, but you seem to want to.
That makes me happy to hear that because I figured that SMF 3 was going to be an update of 2.1.x
I didn't know it was a complete rewrite. And I will gladly give it plenty of time.

QuoteSeems to me that you're not looking for answers, you're trying to have your cake and eat it. You can't completely churn out legacy code every generation, because you'd never battle harden everything. See also vBulletin - 4.x is a complete rebuild and buggy as sin, 5.x is going EXACTLY the same way. Would you rather have mature, proven and stable code or that sort of mess?
That's why I don't use vBulletin: I use SMF. And I know you can't curn out ALL legacy code in every new release, but you could update at least part of it. Looking at the SMF 2.1 forum, it seems that they've upgraded the theme A LOT and added some very nice features that used to be mods. But other than that, I don't see much difference.
Granted, I never compared the code: but I'm guessing there's still code from pre-smf in there that is close to it's due-date. (hence, SMF 3 re-write most likely).

QuoteAnd MyBB is still a confusing mess to operate. Funny how, if you notice, SMF users don't preach to MyBB about how good SMF is, they generally accept that others have their own opinions, but MyBB users sneer down at anyone who isn't using MyBB. That alone, for me, is enough reason not to move.

Mind you, if you want to take the security angle, I found it hilarious when the 1.6.4 bug came along because the exploit wasn't merely an exploit of a vulnerability, it was an exploit of insecurity by design, and I saw how quickly everyone tried to argue that it wasn't.
That's partially what I'm getting at. He likes to constantly bring up the fact that MyBB is better than SMF in this way and that. I do use the fact that SMF has a MUCH easier to operate/navigate admin panel (that is ACTUALLY PART OF THE FORUM ITSELF) to quell many of the debates.
And I never knew about that MyBB security issue because I tend not to care about software I don't normally use, unless it can provide me with a better service.

Quote
I would note, there is an SMF fork out there called Wedge. In over two years that hasn't gotten to an alpha either, despite doing a lot of overhauling of code and pruning legacy stuff and there are still tens of thousands of lines of code barely touched in that time.
A forum is a mess of complex code: that's understandable. But what you said makes me think of something else as well...
Given the several teams of developers over many years, just how "neat" is the SMF code? I've changed a bit of the code here and there to do things but never paid attention much.
I have noticed that some of the names don't make much since. Maybe this contributes to the long development cycle? Sorting through the code I mean.

Please note that it's 11:21 pm so I am not completely sure if I've said anything rude. I do love SMF a lot and I'm just starting to try and integrate into the SMF community.

Daktyl198

Thanks for the replies Antechinus and Arantor!

If I didn't make myself clear, I know there isn't a release schedule because there is NEVER a release schedule for software. At least, not anything I use...
Except linux, but that's in a different category.

Antechinus

Quote from: Daktyl198 on October 17, 2012, 11:24:33 PMLooking at the SMF 2.1 forum, it seems that they've upgraded the theme A LOT and added some very nice features that used to be mods. But other than that, I don't see much difference.
Granted, I never compared the code: but I'm guessing there's still code from pre-smf in there that is close to it's due-date. (hence, SMF 3 re-write most likely).

You're close to the mark there. Yes, the SMF team is aware that the codebase really needs to be completely rewritten at some point to future-proof it. That's not news. However, in the meantime an update of 2.0.x is going to be very useful to a lot of people. Hell, it'll be useful to me. That's the really imprtant part. :D

Anyway, 2.1 is said update, and has been specifically sorted to be an update that can be made stable PDQ (as these things go). So, the existing codebase (including Sources, templates, css and js) has been cleaned up where possible, and some very useful features have been carefully chosen as additions. Bear in mind that praciticality dictates that IE8 still has to be supported by 2.1 (although IE6 and 7 can go rot).


QuoteThat's partially what I'm getting at. He likes to constantly bring up the fact that MyBB is better than SMF in this way and that.

A lot of this sort of thing comes down to personal preference. For every argument that SMF is better in some way, you can probably make an argument that Software X is better in some other way. Basically, use whatever you like most. If your friend likes MyBB, that's his prerogative.


QuoteGiven the several teams of developers over many years, just how "neat" is the SMF code?

As you would probably expect, it's a bit of a mixed bag. Some bits are really good. Other bits are not so good. I think this tends to be the case with most established projects. There is always something you would rather do differently, but sometimes it has to be put off just so a stable product can be released. That's what most users want. They don't know or care about the code that makes it run. :)

Antechinus

Quote from: Daktyl198 on October 17, 2012, 11:24:33 PMGranted, I have found more than one that I like very much (usually from Crip). But in between those I found many that just changed a few pictures from the default theme. It left me wondering about how many customizations the current SMF allows for themes.

Oh, forgot this. The bottom line is that SMF allows a lot of scope, but it does rely on the themer having a good knowledge of coding. Back when the markup and css for 2.0.x was being written, it was hoped that the greater flexibility of the semantic (mostly :D ) markup and the more comprehensive use of CSS (as opposed to tables and inline styles) would encourage more versatility in theming.

The problem is that although you can do an amazing amount just with CSS tweaks, you do have to know how to use structural CSS. Most people don't. They can change colours and images, but when it comes to cross-browser debugging of structural changes they get a bit lost. Add to that the fact that there are limits to CSS, and at some point hacking templates will become necessary, which means you need PHP knowledge as well if you want to do something really out of the ordinary.

It's perfectly possible to do something like the 2.1 default theme for 2.0.x. In fact, I view it as just a sensible (more or less) reworking of the 2.0.x default. There are some quite innovative themes around for 2.0.x, but yes they are outnumbered by the "easy option" because there are more people around who can deal with the easy option.

Daktyl198

QuoteAnyway, 2.1 is said update, and has been specifically sorted to be an update that can be made stable PDQ (as these things go). So, the existing codebase (including Sources, templates, css and js) has been cleaned up where possible, and some very useful features have been carefully chosen as additions. Bear in mind that praciticality dictates that IE8 still has to be supported by 2.1 (although IE6 and 7 can go rot).
Ew. IE. >_>
It's nice to know more about 2.1 though. i've searched and searched and searched and the ONLY thread I could find with "news" about it was the alpha1 release.
And given that I'm new to SMF (like, a couple months new) I had no idea how these things worked :P

Quote
A lot of this sort of thing comes down to personal preference. For every argument that SMF is better in some way, you can probably make an argument that Software X is better in some other way. Basically, use whatever you like most. If your friend likes MyBB, that's his prerogative.
That's true....
But SMF is the best  ::)

QuoteAs you would probably expect, it's a bit of a mixed bag. Some bits are really good. Other bits are not so good. I think this tends to be the case with most established projects. There is always something you would rather do differently, but sometimes it has to be put off just so a stable product can be released. That's what most users want. They don't know or care about the code that makes it run. :)
The sad truth about the world...
Me, I'm completely different. I like bleeding edge stuff, not stable stuff (which is probably why I posted this in the first place against my better judgement) because it (USUALLY) contains new, exciting, and better code than previous versions. And I like new and exciting code.
My friends used to make fun of me in Java Programming class because I would constantly update old projects with new code that was organized and optimized. I was forever associated with the term "optimization"...

*ahem* back on topic: I like SMF over other "bleeding edge" forums though because of it's security. If I'm going to run anything online it has to be secure. Anything I do offline can break my computer for all I care.

Antechinus

I understand what you're saying. A lot of people on the SMF team would like to be able to use the latest and greatest code everywhere, but not all of it is well supported. That means it aint always going to be useful or desirable for everyone. I'd totally love it if CSS3 was finalised and fully supported by all browsers. I'd also love a million bucks, and a Ferrari, and a pony for Christmas, and can I have fries with that kthnx?

Daktyl198

Quote from: Antechinus on October 17, 2012, 11:52:27 PM
Oh, forgot this. The bottom line is that SMF allows a lot of scope, but it does rely on the themer having a good knowledge of coding. Back when the markup and css for 2.0.x was being written, it was hoped that the greater flexibility of the semantic (mostly :D ) markup and the more comprehensive use of CSS (as opposed to tables and inline styles) would encourage more versatility in theming.

The problem is that although you can do an amazing amount just with CSS tweaks, you do have to know how to use structural CSS. Most people don't. They can change colours and images, but when it comes to cross-browser debugging of structural changes they get a bit lost. Add to that the fact that there are limits to CSS, and at some point hacking templates will become necessary, which means you need PHP knowledge as well if you want to do something really out of the ordinary.

It's perfectly possible to do something like the 2.1 default theme for 2.0.x. In fact, I view it as just a sensible (more or less) reworking of the 2.0.x default. There are some quite innovative themes around for 2.0.x, but yes they are outnumbered by the "easy option" because there are more people around who can deal with the easy option.

I'm the same way with CSS (I can do some stuff, but my "stuctural" css is a little wobbly :P )
But I did notice the similarities between the 2.0 and 2.1 themes. I think that's the best way to go about it: update it visually and somewhat structurally, but keep the general look the same.

Daktyl198

Quote from: Antechinus on October 18, 2012, 12:03:55 AM
I understand what you're saying. A lot of people on the SMF team would like to be able to use the latest and greatest code everywhere, but not all of it is well supported. That means it aint always going to be useful or desirable for everyone. I'd totally love it if CSS3 was finalised and fully supported by all browsers. I'd also love a million bucks, and a Ferrari, and a pony for Christmas, and can I have fries with that kthnx?

Ahh. CSS3. And HTML5. I would like some of it's functions to be available to use everywhere.

Antechinus

Back when I first started working with semantic markup and structural CSS, I was grumbling that the bloody stuff was made out of springs, elastic and dynamite. You get used to it though, the more you work with it. After a while you get to know the most likely problems and solutions for them.

The similarities were deliberate. It is an SMF default, after all. Can't have flying lizards and vampire chicks in black leather all over it. The idea was just to cleanup a lot of the crap that wasn't needed, both in terms of code and looks, and revise a bit of the functionality where appropriate.


Quote from: Daktyl198 on October 18, 2012, 12:10:33 AMAhh. CSS3. And HTML5. I would like some of it's functions to be available to use everywhere.

2.1 will use CSS3, but not as much as we'd like to use it. HTML5 is a bit problematic too, unfortunately. I'd be half-inclined to use jQuery to append some HTML5 elements to IE8's repertoire, if they are well suported by IE9 and other browsers. Not sure if that will happen though.

Daktyl198

Hey! I happen to love flying lizards and vampire chicks.
In fact, if anybody can make a nice looking one like that I'd be more than willing to pay a bit >.<

I will say one thing for 2.1: The phpinfo() in the admin panel is VERY helpful.

青山 素子

Not wanting to argue, but I wanted to put in a few comments on some things.

Quote from: Daktyl198 on October 17, 2012, 11:24:33 PM
And I know you can't curn out ALL legacy code in every new release, but you could update at least part of it.

If the code works well, does what it needs to do, and isn't using deprecated functions, why should you re-write it? Sure, if it's some painfully complex function that makes debugging difficult and contains likely security issues from the complexity, it's a good candidate to replace. However, if it's solid, why mess that up? You can spend more time doing cool things like improving an API or working on better theme support, or even just having some brain-clearing downtime.

The SMF developers have, in the past, been very good about making sure to account for any issues PHP reports, including warnings and deprecated notices. I don't think the current development team is any less careful.


Quote from: Daktyl198 on October 17, 2012, 11:56:20 PM
My friends used to make fun of me in Java Programming class because I would constantly update old projects with new code that was organized and optimized. I was forever associated with the term "optimization"...

How many new bugs did you introduce doing that? How complex was that code? Did you account for edge cases and non-documented behavior? SMF is a complex system, it has to run reasonably well with many different server configurations and display across many browsers. Some of that "legacy" code is there to fix a specific issue encountered. If you replace it with an "optimized" solution, you probably just eliminated the fixes that came with it. I'm all for cleaning up things, but sometimes "legacy" code is there for a reason, and you need to account for that before you start off changing things to make it fit your definition of "optimized".

Also, I hope you used instrumentation for optimizing to determine where the code actually needed it. (Sometimes, more complex code is actually faster.) I know a few past SMF devs actually went to the trouble to benchmark various similar PHP functions to see if there were any that offered better performance for the cost of re-writing the code to use them. If you don't benchmark and don't have an idea in mind of where you need to optimize, you're wasting resources that could be better spent elsewhere.


Quote from: Antechinus on October 18, 2012, 12:17:59 AM
It is an SMF default, after all. Can't have flying lizards and vampire chicks in black leather all over it.

What about catgirls? Maybe at least as an easter egg?
Motoko-chan
Director, Simple Machines

Note: Unless otherwise stated, my posts are not representative of any official position or opinion of Simple Machines.


Oldiesmann

HTML5, while fun, is just too risky at this point - browser support still varies widely, the most popular browser is one of the worst in terms of HTML5 support, people stuck on older versions of said browser, etc.

For some real fun in that area though, go to html5test.com and poke around for a bit. You can see the average score for various browsers, see how browsers stack up in terms of support for various things, etc.

The other problem is of course that it's still a draft, meaning things can and possibly will change (though it's been a draft for so long now that we're hopefully getting closer to a final recommendation).
Michael Eshom
Christian Metal Fans

Daktyl198

It seems I started something unnecessary v.v
Sorry.

As for optimization and replacing old code with new code:
NOT ALL CODE IS LEGACY CODE even though it may be OLD. I understand this.
I mean hell, if/else statements were some of the first types of code ever written (in languages above assembly) and it's still around because it's the perfect way to do what it does.
Same goes for complex code. Just because you can do something with a single function doesn't mean that it's faster than a 20-50 line code block.
E.g: ' ' vs " " in PHP. ' '.$variable is faster to run than "$variable" even though it is a bit more complex to write out; especially for long strings.

When I used to "optimize" my programs, I would always write it with a timer built into the code to output how long it took to run the code. I would then test it several times on at least 2 computers.

Do keep in mind that this was a high school programming class: optimizing meant replacing a group of ints or a repeating string with arrays and for-loops.


That's why I added HTML5 to the impossible wish list ;)
And thank you for not mentioning the browser name ^.^
That site is fun. My browser scored a 372 (+10 bonus points). Much better than the last time I glanced at it.

lastly:    青山 素子 I really like your picture, though I forgot her name...
I never was able to finish watching that show (stupid DVD's were scratched)

青山 素子

Quote from: Daktyl198 on October 18, 2012, 03:48:06 AM
It seems I started something unnecessary v.v
Sorry.

It's okay. Keep in mind that many of the people that will respond to topics like this tend to be code-focused and so will make an effort to be precise, especially when discussing code itself.


Quote from: Daktyl198 on October 18, 2012, 03:48:06 AM
When I used to "optimize" my programs, I would always write it with a timer built into the code to output how long it took to run the code. I would then test it several times on at least 2 computers.

That's pretty basic benchmarking, but it counts. Good for you actually doing objective testing, many people don't. If you still like doing that kind of thing, there are some really nice tools out there that will profile full script runs and which work great to show what functions get hit the most often.


Quote from: Daktyl198 on October 18, 2012, 03:48:06 AM
青山 素子 I really like your picture, though I forgot her name...
I never was able to finish watching that show (stupid DVD's were scratched)

It's Yami-chan (also Golden Darkness, 金色の闇, or Konjiki no Yami) from To Love-Ru. Specifically, it's from one of the manga chapter title pages, as Yami never dressed up as a catgirl in-story. I've only read the manga, haven't seen the anime.
Motoko-chan
Director, Simple Machines

Note: Unless otherwise stated, my posts are not representative of any official position or opinion of Simple Machines.


Daktyl198

Ah yeah. I read the first series of To-Love-Ru all the way, but I haven't gone past that.

I thought she was also in Black Cat?

青山 素子

Quote from: Daktyl198 on October 18, 2012, 05:50:47 PM
I thought she was also in Black Cat?

Not to my knowledge, although both series had the same illustrator (the author of To Love-Ru was someone else) so it might just be a design familiarity. Then again, there might have been a cameo. I haven't read Black Cat.
Motoko-chan
Director, Simple Machines

Note: Unless otherwise stated, my posts are not representative of any official position or opinion of Simple Machines.


Daktyl198

Ignoring the fact that this discussion has gone horridly off topic: I was mistaken. The designs are noticeably different. I guess I just haven't read/watched either in a while :/

Advertisement: