Wedge

So...

- All files transferred. I also moved /avatars to /assets, for good mesure.
- Moved temp repo to my local website, so that I could test. Crashes in a funny way.
- Fixed variables, etc. It's running.
- Theme is still completely broken. It's going through wetem::hide() for some reason, I'll look into it.
- What have I got into, really..?!?! This is AGONY. The amount of code just to account for theme handling is incredibly huge. I think my estimate of 2 weeks is going to be correct, but it won't be 2 fun weeks.
- I may do this conversion in two phases. The first phase would keep all theme terminology, and some of the variables, and the second phase would attempt to convert everything.

Okay, so after 8 votes, still none for keeping themes in, so I'll start work on it.
I'm just hoping that it doesn't screw up everything when it comes to moving folders around, of course...All right, I'm not crazy then! ;)
I checked it on mobile, and the fixed header is making it hard to appreciate the exprience. It has too many entries, so it spans over four lines, and makes the header twice as big as its background, meaning it overflows. You should disable the fixed positioning for it in narrow viewports. ;)That's one of the problems I had with my Wedge implementation, yes. Plus, the fact that due to soft-merging, it's more complicated for me to add the link outside the message block, and adding it inside would screw up the layout, as most of these divs are positioned in a way or another... Well, gotta say, Wedge topics look great, but they sure have a few hacks that make it harder, I guess, to restyle completely... (You can still move the avatar to the right side or whatever, like Wine does, but of course it's not a hard one.)I stopped The Blacklist after a couple of episodes (I love James Spader, but he seems to be re-playing his Boston Legal character, only with a dark side... I'll probably catch up when there's a half-season or full season available, but as a weekly show, it doesn't excite me). Person of Interest, I always found the concept to be silly, but Once upon a time had a sillier premise, and it turned out to be great (well... Until the latter episodes in season 2, of course. It's only starting to get interesting again these days, and even then, it's so full of plot holes, I wonder where the plot went... Certainly not to Neverland.) What I mean is, I watched 5 episodes and I was like, "meh." But I heard that Amy Acker is a regular in later episodes, so it might be enough of an incentive to watch. (I totally dig her. Sue me. She's a modern Audrey Hepburn.)

Personally, my unmissable weekly shows: The Legend of Korra (a true wonder, season 2 is awesome), Homeland, The Walking Dead, OUAT, Quite interesting XL, Elementary, South Park, The Big Bang Theory and Misfits. And Community is coming back soon, so add that to the list. And Doctor Who, and Sherlock. And on summers, Wilfred.
The rest, that I watch when I'm bored: Agents of SHIELD, How I met your mother (season 9 is the worst ever... Too bad), Mom, OUAT in Wonderland (it's even sillier than its parent show, can you believe that?), and older shows I'm catching up with: Mike & Molly, Defiance (such a great pilot, such a disappointingly slow show afterwards), The Almighty Johnsons (I think I read something from Aaron saying it rocked, so I started it, and indeed it rocks, but I'm holding up for subtitles of season 3), and I only dropped one show with no redeeming chances: The Crazy Ones. Robin Williams, I love him but I prefer him when he doesn't play crazy. I'm also considering dropping Toast of London. I like the actor, but he's not THAT funny.
Recently, I watched Dirk Gently. That was fantastic. Too bad it got cancelled. Same for Spy, a fun, mindless sitcom. At least it got sort of an ending.

Anything else going on these days, that I might have missed..?That's an interesting theory about Quinn. I doubt it's true, though. But it would explain things.
As for the other scene, I think it's not surprising at all. His wife obviously came back to help him in a rough situation, as a friend, not because she wanted to give him another chance. Saul is obsessed with work. He pretty much knew there was a price to pay for her support. That's the price in question. But he's a child when it comes to relationships, so we can't know how he'll deal with it. Or if he will.

rev 2318
 6 files changed, 103 insertions(+), 36 deletions(-), 2.97 KiB

+ Added support for relative dates in topic pages. (topic.js, index.language.php)

@ Need to know: this isn't going to stay 'as is'. I'm only trying to clean up my uncommitted stuff, and this is an important part of it. Also, an issue: done always at runtime, so you might see the absolute date at load time, a second before it becomes relative. As I said-- won't stay like this.

* Forcing relative dates to be absolute after 2 years. (Msg.template.php)

+ Added a helper function, time_tag(), which will be helpful to add these tags outside of topic pages later. Also removed a couple of lines that didn't make sense in forum_time(). (Subs.php, Msg.template.php)

* Moved runtime code like code+flexbox fixing, follow_me() adjuster and relative dates, to a new function that will get re-called whenever necessary. (topic.js)

rev 2317 -- yeah, ultra-short one, but I couldn't stand having it unpushed.
 1 file changed, 1 insertions(+), 1 deletions(-), 503 bytes

! Repeat a 1000 times: do not commit any last-minute optimizations without testing them... This one made we::is('mobile') return true on desktop browsers, which was... Not so good. (Class-System.php)

Hey, it's starting to look like I'm gonna have to get to work soon on removing theme support... Seven votes against zero in favor of doing that! :):)Apologies for implying otherwise, but I just saw your desktop version, and it's looking great.
So far, I'd only tested on my Android phone, and it didn't look as good. (Plus, it's very buggy in home pages.)
The only thing I'd seriously rework is Display pages, mainly avatars and page indexes (these circles, they have a 'meh' effect on me.)
The rest looks great.
Oh, also, when I saw your fixed header, I thought, "he must have run into the same problem that I did...", and you sure did. For instance, if you click a link to a specific post in a topic (not a topic itself), the resulting page half-hides the post, because it needs to be straightened by JavaScript into moving the page further up so you can see the desired anchor. I'd started to do that myself, and then I realized refreshed the page complicated matters: do I need to execute such JS in that situation? Depends on whether the browser forces the anchor position on refresh, and not all browsers agree on that, so... It gets complicated.It's your most common phase, and the one I'm most jealous of. ;) (cf. Minimalism, Simplicity, ViennaBBS, and now Protendo.)
Really: I can't remove anything from Wedge without feeling I've removed too much, and I don't want a root skin that has too much CSS. What I want is a minimalist root skin, and then a 'full-featured skin' with styling everywhere, and give skins the ability to choose either of those as their parent.Well you could always play with the HTML if you think you can make it more flexible for skinners. ;)Too many classes tend to make the CSS files bloated, though.
I kind of miss the days when SMF 1 had a single CSS file that weighed around 10KB, but hey... That's just me ;) I know that this particular SMF was also very, very reliant on tables and inline styles.AFAIK, you've always had a read access to the SVN at arantor.repositoryhosting.com, didn't you use it back in the day? ;)
Apart from that, that repo is now a month old, all development is now happening at the Bitbucket repo; still a private repo, so if you want access to it, feel free to ask. (I still have a slot open for my private repos, but it pretty much requires me to send you an invite by e-mail and have you create an account especially for that, following the link in the invite. That way, I get my last allocated 'promotional' slot.)Yes. But I could very well release a new alpha right now, as it's relatively stable. I'm just waiting for the theme system to be removed, because it implies restructuring all folders, and I don't want to tell people to do it manually on their side ;)I've never required of anyone with repo access to give regular feedback; knowing you're in is enough for me. I have a limited number of slots, though, so if someone doesn't seem to share anything (whether feedback or code or anything really), I may have to ask them if I can remove them from the list. OTOH, I could also add as many users as I want, as right now no one but me has commit access to the repo, and I think Bitbucket has an (hopefully) unlimited number of private users, as long as the repo is read-only.

Yes.He's... Complex. Even in the days of Wedge, I didn't understand him for all he did.
I mean, these days, I'm starting to wonder if his goal with his work on Wedge was the same as mine -- making the best forum software we could think of, something we'd want to use ourselves, something without the dead weight of the many mods and themes in SMF that always need to be catered for, something without the 'mass' of users in mind, targeted to people who're willing to get to work to have the best software around, and who don't care whether the software they use has a large user base, recognition or mod/theme library. It's now looking like his goal was simply to make a prototype of SMF 3.0... Because it's certainly what he's working on, these days. Taking his 'best' of Wedge and putting it into SMF.

Of course, there are implications to this line of work, that make me chuckle. I won't elaborate, though.

Arantor is what he is. He posted tens of thousands of messages on sm.org, which to me always sounded like "okay, I hate the SMF team, but I can't stop myself from posting all day long, so I have to have somewhere to post all my knowledge."
It made it pretty much impossible to follow him on the myriads of forums he was registered on.
So, he would start discussing things that I had no idea what they were about, and I would just let it slide. Things like this post here... Seriously, why did he care about an incompetent developer at the SMF team..? I mean, he WASN'T going to be using SMF any time soon, was he...? Turned out he had so much pride in Wedge, he used SMF to build his latest website. To me, that felt like a public renouncement. It was only confimed when he came back from his trip.I think we all did. Including me. I'd already, ahem, 'experienced' the difficulties of trying to be frank with him. I learned my lesson, and progressively stopped sharing my thoughts on what he made in Wedge that I didn't like. I'm somehow glad I know I can remove them now, but I also realize it would take time to revert some of his 'innovations', and I'd rather spend my time finishing up a usable beta.

He was just hard to work with. He had an answer for everything, and he gave them with so much aplomb that you just couldn't discuss them with him. Where I would post poll asking people where they think I had a good idea here or there, he just didn't bother -- his ideas were for the best, after all. Well, a majority of them were, but when you start fearing that he'll snap if you question one of his future implementations (or, worse, that he'll start arguing over it in novella-sized posts that no one can bother to read, let alone reply to), something's really wrong.

I for one, seriously hope that people will keep telling me what I'm doing wrong. Wrong direction? It's all right, I'd rather know when I just got started on something, rather than at the end when someone dares to speak. Perhaps I'm more understanding of 'newbies', simply because I'm a strong believer in the fact that "you're always someone else's dummy". In whatever field you feel you excel at, there's always going to be someone who knows better than you. I realize that (I even realize that person is often Arantor!), but if I keep thinking, "I'm not the right person for the job", then I'll never do anything. My Mahjongg Solitaire game sucked when I started it, which is why I released it for free, which is why it got popular, which is why I kept working on it, and years later made it into something that still to this day tell me is a masterpiece. I still think it was a piece of crap, but you have to draw a line: yes, someone else could have done a better job at writing Kyodai Mahjongg, but he probably wouldn't have bothered to work on it for 9 years, like I did.
When you have the guts and the will to do it, and to finish it, then you can do it. Doesn't matter if you have to restart from scratch several times. It will take what it takes to do it.

TL:DR. Yes, he's hard to see through, and whenever you start talking with him, you can't help but feel frightened that he'll end up shunning you. Most of the times it would end happily and he'd be cordial and nice, but do one single mistake, or insist a bit too much for his taste, and he'll crush you. And now where he is, I won't be there to possibly help smooth things between him and other people.Yeah, it still needs work, but it's something that'll end up as one of Wedge's finest features.
I mean, some of my ideas aren't implemented yet, but... You'll see! Once I'm done with the contact list system, I'll devote myself to privacy settings, and the bulk of the work is done, so you'll see.

Interestingly, I caught a topic where a certain Arantor told someone that "SMF will never (stoop so low as to...) implement privacy features, because it's not what a forum is for." Aside from the fact that that particular topic shows his 'style' (half arrogant, half trustable), as well as his 'badge' thing (yes, we know you're proud to be at SMF, even though you said the exact opposite in this very topic), he also mentions the disastrous impact on privacy... Which, as championed recently by Pandos, is not exactly THAT big a deal.[1] So... Yeah, whatever.Well, we did sign up knowingly, you see... We just didn't expect to have more than six months of work ahead of us, which was manageable. My plans were to simply import my SMF->Noisen.com patch to Wedge (around 900KB of code which I progressively reduced to 400KB through implementations and deletions, before I finally gave up and figured the rest would best be rewritten), and his plans, err... He had more ambition than I did, which is the reason why we postponed every six months, and consequently why I started implementing other things 'because we had time'...

Now that I'm alone on the project (as the decision maker, I mean), I certainly can decide when I'm going to feature-freeze the entire thing. Not right now, but it's getting close. Hopefully, next month I'll have some releasable beta. And if any of Arantor's code is in an "unreleasable state", then we'll just work on it. At worst, I'll simply remove said code.[2]Well, if I ever was going to have people pay for Wedge, it would be for an inexpensive price, less than $50. Probably around $30. I wouldn't expect to sell more than a few hundred copies (at best!!), so it wouldn't even pay for a month's worth of work on my current 'rate', so... I don't know why I should bother. If I could make it work as a sustainable revenue maker, then I certainly could envision working on Wedge for years and years to come. But alas, I'm no marketing guy, and I have no plans to turn Wedge into a marketing machine. As I said before, I'm only targeting people who know a bit of PHP and are willing to do what needs to be done for Wedge to work well on their setup. It'll mean less support questions for me, and happier power users.

PS: oops, I just realized it's an Arantor-length post. Sorry about that! I could probably shorten it by half, but I'm not sure it's worth it.

1.

Actually, to be fair, Pete did tell me in private that he agreed to a suggestion of mine that we could simply disregard topic privacy at query time, and just hide unreachable topics at display time, to save processing time at the cost of getting empty pages if you stumble upon a series of hidden topics. I'm not going to do it this way, though; I'll only do it if privacy can provably be shown as having a real impact on overall forum performance.

2.

It'd be interesting to see which Wedge would be best: a Wedge without my code, or a Wedge without his. They would still both be very good, I have to say!

Yup... As I said, since I was able to reproduce on my iPod Touch with iOS 5.1, from now on I won't have to rely on others to tell me about whether it's okay or not.

Unfortunately, as I pointed out yesterday, I'm in a (sad) position where I have to consider disabling user menus entirely for iOS users.