Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Topics - Arantor
The Pub / Not So Mixed Signals
« on September 2nd, 2013, 03:50 PM »
Well, Nao wanted a topic, so here we go.
I've never had any problems with writing code that you didn't like, and reverting it...
I don't ever remember this actually happening? On the other hand I can think of several examples where you changed things, I made it very clear I didn't like it, explained why and was completely ignored. Like the sidebar changes in recent times - so much so that the *need* to move PMs out of the menu was apparent because it was the only way to know that I'd have messages (since the PM popup doesn't really count, especially if the PM count gets out of sync, which it does occasionally)
I see your leaving as an attempt to kill the project. Because you know I'm not comfortable with being the only one to manage it. There are so many areas that I entirely trusted to you, and now I have to learn about them thoroughly-- or just give up about them.
It'll be good for you to develop your skills, actually. Having at least an awareness of security, of optimisation of DB stuff (which, by comparison, is far more of a killer than byte optimisations tend to be), will make you a more rounded developer.

There are so many areas I trusted to you too, but when I had concerns, very often they weren't listened to. And when I did so much as breathe in the direction of those areas, I felt like I was always doing it wrong because every time you'd rewrite the code (how often, seriously, did I rewrite your code?)
It's not a good feeling, I'll tell you. Also, you were always the one who said Wedge wasn't in a releasable state. Sure it is.
Then release it and let the users be the judge.
Perhaps you were just looking for a way out, I don't know... But I have to deal with that now.
I have more respect for you than that. If I wanted out, I'd just say so. You even told me that I owed Wedge nothing and could walk away at any time. But it wasn't about 'wanting out' as such. The problem wasn't the fact that Wedge is a black hole where thousands of hours of work go into and for seemingly no reward, the problem is that I began to actively hate the experience of developing for it and I resented putting my time into something that made me feel like that.

I don't know if you've had comments from community members about my attitude and expressing sympathy or not, I daresay you probably have. On the other hand, I've had some comments (from people that, in a shock and surprise move, I'm choosing not to name) that they're surprised we ever hooked up in the first place, because they knew it wouldn't work out (and they're surprised it lasted as long as it did)
Anyway, I will release it alone, but you didn't make things easier for me.
Yes, because it's all about you, isn't it? On the contrary, I made things easier in several ways - firstly, I cleaned up the licensing situation as best I could. Secondly, I made sure that my own pet project which may or may not go anywhere doesn't have Aeva in it, so it's not like Wedge is losing uniqueness, and thirdly, I said it indirectly but might as well make it more formal: you're absolutely free to use any of my contributions in any fashion you like. I'm hereby granting an irrevocable licence to Wedge (or whatever the project becomes called) for any contributions I made, code, documentation, anything, under the BSD licence. If you want to relicence it, fine. But it's there.

If I wanted to kill the project, there are ways I could have done that, and far more effectively. I could have totally stitched you up, because I could have made the threat that nearly killed SMF in 2010 and I would have had just as much legal right to do, because I could have demanded all my contributions be removed because I still hold my copyright to them, which would have left you in a very strange and awkward position, as wesql would have had to be removed or at least reverted back to $smcFunc, along with removing the warning system, the ban system, the plugin system, the language editor, the rewrites of the bbc parser to use the database... the list goes on.

The bottom line is that I finally had enough of dealing with your attitude and I can't do it any more. Even now I feel like I'm forced to justify my decisions to you or explain them - I don't owe you explanations. I certainly don't owe you justifications. But here we are.

As for why I didn't make a topic, it's because it was a couple of lines of throwaway comment that didn't need to be a topic, more a simple observation that got sidetracked and blown entirely out of proportion yet again.

My pet project is experimental, it may not go anywhere but while I'm concentrating on C#, I'd like to feel that the last 10 years honing PHP and 3 years on Wedge wasn't a total waste.[1]

Will my pet project be competition? Not really. It's more really poor man's competition for WordPress than Wedge, but you believe what you want to believe, you usually do.
 1. Rule 5: Don't waste good.
Since things boiled over last night and I quit doing the dev thing, I set my PMs to admin only. And yet I have two PMs now, not from administrators or contacts.

Just thought I should report it, in case it gets overlooked some more.

Sorry to trouble you further.
If you try to search in the language editor, plugins whose language files aren't in their top level folder (where plugin-info.xml is) throw errors as well as not being added to the results list.
Off-topic / This is why I complain about tracking hits
« on August 19th, 2013, 01:30 AM »
No-one seems to understand why I complain as much as I do about the fallacy of maintaining a hit count.

Well, here's a perfect example. We have 9 users online. 2 are real, 6 are already logged as bots, the last one is... shock horror... a bot too.

Yup, that means that 77% of the users online when I just looked at not actual users. I don't think I've seen it go below 50% in the last few weeks (whenever I've been able to look)

Since we can't always keep an eye out for bots, and can't always keep the list up to date, seriously can't we just stop counting the number of online people yet? Or at the very least stop counting the records? (I still don't even want them shown to admins, to be honest, I'd rather just not track them)
Other software / Arguments about the credits
« on August 16th, 2013, 03:07 AM »
This is petty but I'm frustrated, having just been part of an argument about the SMF credits.

They're already messed up and in 2.1 they're actually likely to get worse. Right now, for example, the plan is to have the current team up first, followed by everyone else as 'Friends', followed by Github contributors and stuff like that.

Now, I think it's nice they're actually getting Github contributors. I also understand there are a number of people who don't want even the Github contributors credited. There's nothing better than that for 'we'll take your commits but fuck you for wanting some credit to them'.

But the bit that's getting me frustrated is the absolute inability of some people to understand why lumping all the Friends together is actually fucked up.

Let me put this straight, and I'll try to use small words so everyone understands me.

1. SMF is a software project. Without the developers developing it, there would be no SMF.
2. If there is no SMF, there is no need for support, customizers, translators etc.
3. The only conclusion I can then draw from this is that developers are actually necessary to the project in a way that other roles aren't.
4. Support, Customizers, Translators are all important. But they, by necessity, must come second to developers because without developers, there's nothing to support, nothing to customize, nothing to translate.

Assuming we agree on the above:
a) Developers' contributions do not stop being important just because they left the team. There are still bits of the original SMF 1.0 (and, while we're at it, elements from YaBB) still in 2.0 and even in 2.1.
b) The current developers, as fine a people as they are, did not write 95% of SMF's code, because of the above.
c) Why, then, are the current developers considered as rock stars compared to the 'Friends' who are no more important than anyone else despite the fact that some of the Friends actually wrote the damn software?

It seems very simple to me: anyone who actually contributed code to SMF, i.e. those who MADE SMF need a higher billing than the people who support it etc.

In case anyone thinks this is about my getting my name higher up... no. I never made the dev team for various reasons. My code contributions are small enough that I don't personally consider them important enough to be considered a developer. I have no qualms with, then, remaining in the Friends list, provided that the people who actually made SMF get a slightly higher billing - after all, they're the ones who made it, and if SMF hadn't been made, I wouldn't have had anything to contribute.

And yet this mindset is still rife in SMF, that developers are not any more important than anyone else. Yes, that's right... in a project based around a piece of software, the people who make the software are not particularly important.

Imagine if that applied to other industries. Imagine if we told book authors, that suddenly they have to credit the publisher as being as important in the book's making as themselves. Now, books have the author in big letters on the front, and the publisher's logo on there too but when was the last time you saw the publisher receiving equal prominence to the author? What about the typesetters? The proof readers? The author's best friend who proof-read an early script? The author's partner who gave them moral support? By SMF's definition, we should be giving all these people equal credit to the author, regardless of anything else.

Apparently I appear to be in the minority of people who understand this. It's sad really, especially in light of other recent events where for one brief shining moment I was considering helping with SMF core development.
Features / Improving search
« on August 9th, 2013, 07:40 AM »
So I've been thinking about searching and the way searching works and I've concluded a number of things.

1. I want Sphinx and ElasticSearch in the core
Both Sphinx and ElasticSearch are pretty mature. Both have live update features now, so there's no reason we can't support them both with a sort of push mechanism (rather than Sphinx's pull mentality, like the old API was geared for)... the API needs rewriting to support either anyway and I might as well do it all together.

2. I want to natively support other types of data than posts.
The current setup doesn't support anything other than posts and I want to natively offer support for other stuff - calendar, helpdesk etc. The backend can support these extra things with some work, and pushing these also allows nice support in both ES and Sphinx.

3. It adds some nice feature parity with other systems without adding a ton of headaches for support.
XenForo offers ES with a $60 plugin, though I'm not entirely sure why. IPB has Sphinx in the core. Neither appears to have a huge support overhead because of them. And for the most part once they're done, they're done from our point of view.

4. The most controversial aspect of this is what I want to propose last: ditching unindexed searching.
Right now, the default searching method in Wedge is as it is in SMF: no index. It's slow, and doesn't scale beyond a few tens of thousands at peak. In fact, where we are right now on is probably about the limit of what we can do with an unindexed DB before performance starts to go nuts. (40k is really the upper limit)

Now, partly this is because we've never configured it to be anything else, and most people just wouldn't know to do so because they wouldn't know any better. Now that's fine, because we know that people don't generally touch the settings unless they're directed - but using the search index would deliver better search performance from about 1k posts and up (and largely a push in performance terms for where things are right now for fresh installs)

I see no reason why 'no index' ever needs to be a valid search type. I'd suggest dropping that entirely and using the 'no index' option to mean 'no searching'. And then leaving the other index types to be actual index types, which would simplify the search code as well (and properly allow for it all to be segregated back to the APIs, some of which has already occurred but plenty more still to do)

This would leave us with three working search types (standard - formally known as custom, ES, Sphinx), of which 'custom' would be set as default on installation and would be populating posts as they are created (rather than having to deal with a huge index creation at once)

ES and Sphinx are both VPS level options, but there's no reason we can't have people pushing content to these indexes while using the custom index - plus of course there are always options for rebuilding indexes.

Does any of this make sense? Any questions?
Off-topic / Food for thought
« on August 8th, 2013, 03:37 AM »
Browsing around the intarwebs I found this article.

It's an interesting argument specifically about Firefox but it does make me want to go back and look at what we're doing...
Archived fixes / Soft merge / ignored user
« on August 1st, 2013, 06:16 AM »
If you ignore a user and they have soft merged posts, both posts are collapsed as they should be but the second (or presumably third and onwards) softly merged post cannot be uncollapsed. What happens is that you click the link, it animates the uncollapse, then it animates recollapsing it. Tested here just now.
The Pub / Posting from mobile, including Aeva
« on July 25th, 2013, 09:30 AM »
OK, so as you guys know, I'm on this crazy trip across the US and we've been logging our stuff on our little website which is based on SMF + Aeva and a custom theme.

There are some interesting things we've found from doing this that I want to share, and since the Wedge version of Aeva is still basically Aeva from a user standpoint, I think a lot of that stuff stands. Note that this is SMF without a mobile theme and not using WAP because there are other things tied into the site's posting page that aren't in WAP and I didn't have time (or energy) to code them up multiple times. (Namely, the site has a topic tagging system and also a facility to geotag individual posts with a location. This requires various splices into the posting setup.)

1) Posting, generally, sucks from a mobile phone. It actually really sucks on an iPhone 4S, I'm actually surprised more people don't complain about how bad the experience is. Even with the phone in landscape, getting the subject and then the text in is painful. We don't really need formatting much in mobile, even the bold etc. is probably too exotic for the most part but perhaps a collapsed menu of some kind to bring up the tags would be OK. Ideally though I'd consider doing what Tapatalk does: the post interface for a new topic is little more than two text boxes. (Of course I need other stuff but that's another matter entirely and the specific requirements we have for Crossing Overland will actually necessitate a custom app to support background data being gathered and batch uploaded to the site later on.)

2) Does Wedge's version of Aeva support spitting out the non Flash version of embed code where we have it? I'm thinking YouTube in particular. (I can't exactly do a hotpatch on the road to splice the code in, but there's at least one video on CO that uses the standard embed code.)

3) Media uploading can typically be much, much simpler than what we currently have. Sure, most of those options will be useful for some people but for the most part, especially when uploading from mobile, it's just not necessary. Perhaps a collapsible menu for the stuff like album covers and the extended information.

4) We really need a mass upload option of some kind (even just AJAX uploading a la Dragooon's plugin for regular posting) that works without Flash so that uploading from mobile is not a hideous chore. It's gotten to the point where we've broken out the one laptop we did actually bring to handle mass uploading, but that's a 2006 MacBook and doesn't exactly handle it *well*. (In unrelated news, the camera connector kit for iPad works really well)... actually, AJAX uploading for media would truly rule just to have it anyway, mobile or not.

5) The selector bit for media items where you can copy the bbcode is not really copyable on a mobile device and for reasons I can't entirely figure out, the item doesn't usually come up with anything useful in there other than the id number.

For example, we uploaded a picture, and the item in question just had:
Code: [Select]
[smg id=216]

as the text. Which would normally be fine but we wanted:
Code: [Select]
[smg id=216 type=preview align=center caption="First sight of the Pacific"]

or similar and this is a real pain to add on mobile if you want kind of nice formatting on blog posts. We find ourselves copy/pasting all the time to make this work... I don't know about on desktop yet but on mobile at least it would probably be neat to have the media button not just go straight to an upload button but instead to a media browser from which you can upload if you want, and then have the media browser cover the gallery contents. This would mean we could just upload everything, then filch it from the gallery easily, or upload one at a time without any real fuss either (since the popup could handle the upload AJAXively which would be slick even on an iPad). Fairly sure I've mentioned this before but IPB actually does it this way, and thinking about it, it would also make mass upload much nicer for posting images to the site and then pulling that into the post and that would greatly smooth off our workflow - and I'm fairly sure it would for others too.

Aeva works truly magnificently as a storage ground for media items but I gotta say, the workflow of getting items into posts isn't ideal and it gets worse with the more items you add and until recently I'd pretty much used it as a storage ground, rather than making it part of a cohesive blog structure.

6) Can we actually do something with the textbox sizing and positioning? What we've found with an iPhone (it's largely a non issue with iPad) is that there is some crazy resizing crap going on that means you tap to select the textbox and then it zooms in to the point where the textbox is larger than the screen. Now, I'm sure this is at least partly no longer an issue because it's using SMF 2 code rather than our code and I can't really test it on Wedge right now to be sure, but the way it does zooming on the fixed-width container is really frustrating since it means to type in a post you have to swipe left and right to actually see everything, and even composing the line-blank-line-blank-line posts I've been doing for the blog for the live video updates (yes that's me doing it on Louis' iPhone while he's driving) is frustrating. Again I'm thinking like Tapatalk here, where the box is sized for the screen and it all fits without any scrolling horizontally or any such nonsense. I would kind of presume @Dragooon's mobile theme for SMF works nicely in this regard but again right now I can't really test it. I should note this is the first time I've *really* tried to use an SMF setup for posting and even though I know we have a slightly weird theme, I can't be the only person who finds the experience of posting even 3 lines of text with 2 blank lines tiresome.

I really don't know how much of this stuff can be fixed by way of a mobile skin. Seems to me there is far more at stake than simple CSS resizing (e.g. pulling a different editor component for mobile that doesn't just push all the buttons as normal but has a totally different presentation and code to support it) and some of it will need much greater change if it is to be facilitated, especially the uploading workflow stuff. (Btw, I'd like to see the AJAX uploading of attachments as a core feature just in general. Would be easier to support it in Aeva too if the support files are already present.)
Other software / Powered by SMF footer
« on July 12th, 2013, 09:43 AM »
Something I've been noticing for a while, but it's only recently that I've had enough information to form any kind of hypothesis about it.

The biggest way of sites getting found is by having the 'Powered by SMF' in the footer. Not a specific version but simply that footer. I have a number of SMF sites that have questions that haven't been changed in months - but that the bots just haven't found the sites. All the sites are in Google but they're not being picked up by the spam bots at all.

It's one more trait to discouraging spam, giving them something less to target. The other software has essentially the same problem: if you have powered-by-whatever in the footer that's common to most installs, it's an easy way to be found.

Of course, this is at odds with the implied advertising of software where the footer is used to indicate what the software is that the site is running in order to get some cheap promotion, but perhaps it's reason to consider something else instead that can't be searched by Google. I don't know what that would entail at this stage but it's something to think about. Meanwhile I'm quite happy being low-spam ;)
The Pub / Decluttering/rejigging the top menu a little
« on July 12th, 2013, 09:24 AM »
Suggestion, sort of sparked by something Antechinus said on Elk: bringing the profile and PM stuff under one heading...

How about we bring a new top level menu item into play, whose top level heading is the user's own display name? Under it would be the profile stuff, plus the PM stuff and if we wanted to go really crazy we could put the notifications thing there too. Or bring it all together under the notifications popup.

This has a lot of interesting implications that all tie together - I guess the comment I made about the menu bar being hidden on mobile fell into the ether.

So, let's back up. If you're on mobile, you won't see the menu bar because it'll be in the sidebar and most people won't check that because that's how most people are. If they can't see it, they won't check it most of the time simply because if it's not there, it's not relevant to them.[1] And so people who predominantly use mobile won't see that they have personal messages.

Solution 1: send a notification about PMs. It reduces the priority of having messages in the top level menu, there's simply less burning need to have them up there (though I'm not saying they have to go, but they do get to lose some of their otherwise premium status)

Solution 2: pull profile and PMs out of the menu. Pull them up by the notifications area. Maybe even give PMs a similar interface to the notifications one for showing the inbox, and a similar set of controls for configuration and sending a new message. We can also refit the access-to-own-profile from there and make it easier for people to get to their own configurable items. I'm thinking this would ultimately be in line with overhauling the profile anyway.

Solution 2 means we get to separate the profile and messages items which don't really fit in the menu anyway if you stop and think about it, they're just there because there isn't anywhere else for them and it means we have less need to pull the menu out of page layout in mobile because there's less of it and it means what gets up there can be more meaningful to the *site*.

Consider: you get to show: Home, Admin, Media, Members, Logout. (We could even move the login/out options up there, too) Unlike the profile and so on which relate to YOU, these things relate to the SITE. To me this just makes more sense to have as principle navigation. We can still do PMs-getting-notifications and in fact probably should convert it to that if feasible[2] and that would also solve some issues WRT navigation if the menu bar is going to be hidden. (And we get to streamline preferences for PMs, potentially, if not. I dunno. I just think it'd be a nice way to tackle it.)

Pulling this stuff off the menu also means that we don't do weird things like highlighting the profile tab when you're not viewing your own profile which seems semi logical a thing to do but it's also not really so logical when you stop to think about it. The button is highlighted even though you're not actually *there*, on the place you'd be if you'd pressed the button yourself.

As a follow-on to this, I'd consider putting a 'What's new' option or similar on the menu in their place. I'm not sure whether that would be pulling in Unread Posts or Unread Replies or both in some fashion but I'd certainly bring them out of the sidebar to that point.

At this point I'm mindful that the sidebar is increasingly for 'less important content' in which case I'm trying to discourage major navigational items from being hidden - unread posts/unread replies is useful navigation, if not quite as 'major' as perhaps it might be. But if it were on the top level menu, that makes things a bit different.[3]

And now, just because it's that point in time, I'm going to invoke the devil just to reinforce my point.

(click to show/hide)
XenForo does this. IPB does this. Facebook does this. They do essentially what I'm proposing because of the reasons I'm proposing them: you pull stuff that isn't about the site out of the place where site content navigation is and you pull them together. Alerts/notifications and your profile and so on should be out of the way of site content precisely because they're not site content. They're your content and only directly have relevance to you, not to everyone. So they don't belong in the navigation that everyone has.

There is a larger argument here. I'm not advocating copying the competition. I'm advocating taking the same path they did for the same reason they did. Because form follows function. The profile and PM functions are not site content. They shouldn't be with site navigation. If that includes log in/log out navigation too, fine. Just get them the hell out of the main menu because they don't belong there. Make site navigation for the site content, not the user content. Pull that stuff out the way into the top of the page where users will always have access to it (much as they already do, except currently on mobile), we have the room, we already have UI/UX precedent for it with notifications (implicitly a USER function not a SITE function) and language selection (which is totally a user function anyway) up there, let's finish it up with making it consistent.

If it makes you feel any better you can safely ignore the spoiler above and let the arguments stand on their merit. But if you can tolerate my pointing out what the competition does and why they do it and why we should follow a similar path, read it.

@Nao: Random question. I was going to use footnotes in the above spoiler tag but then I realised that footnotes will be pulled out of the spoiler and into the footer list along with the rest. Should footnotes adhere to the same parenting like they do with quotes (i.e. staying inside the quote they were made in?) or have their current behaviour?
 1. This is a killer reason why auto hiding the sidebar is actually a bad idea and if you remember it was the key reason I didn't want the notifications in the sidebar, because being at the bottom means people aren't going to see it. The same argument can and should be made for the principle cross-site navigation: if it's hidden by default, it's a butt ache to navigate around.
 2. Though I'm mindful of the consequences of this in terms of the code if you have PMs going to a bunch of people at once, especially if that's going to include announcements having the same code path. @Dragooon, what are your thoughts on this? How sane/insane is it to try and generate a large number of notifications for large numbers of users at a time? I know in the newsletter code we already have the continue/not-done code so we can stage it a bit better, but I'd rather not generate several queries for each notification if at all possible. But I don't remember how that part works.
 3. See, then, if we kill those two items from the sidebar, we might as well remove the avatar from there too. Time too while we're at it. These things are simply not that useful, doubly so on a sidebar that only some users will even see anyway.
Test board / test
« on June 24th, 2013, 07:46 PM »
! ! ▦ ▦
Archived fixes / Off limits notification?
« on June 21st, 2013, 03:56 AM »
See attached. Somehow the notifier thinks the board/topic is off limits to me - and yet if I click the link it works exactly as expected and shows me the topic, so it's not actually off limits.

Notification id 611 if you want to check the database.

* Arantor is still not seeing email notifications (all set to periodical)
Features / More useless nonsense
« on June 19th, 2013, 05:50 AM »
So, long story short, as part of what I'm doing out here in the US, I'm building a website for it and it's an SMF install[1] with a ported WP theme[2]

And so while I'm gutting SMF's shonky old markup (and dear god it is, I forgot how far we came with it), I got to thinking. Yes, really, all this build up for a small point.[3]

On the board index we have:
Most Online Today: 25. Most Online Ever: 98 (July 7th, 2012, 08:52 AM)

or similar. I want it gone. I see no reason to keep the maximum number of people online today (which is guests + members + bots), nor do I see a need to keep the most online ever stat. They only serve as a 'we were this busy at > this < time'. Which if the record was set some time ago, could be a problem.[4]

I just see no reason to keep it. It's logic we don't need, queries we can save on the board index, etc.
 1. Needs stability, I won't be able to babysit it all the time, nor will the people taking it on after I'm done be as technically minded as I am so maintenance needs to be easy.
 2. Porting a WP theme was easier when I stopped trying to look at the PHP. There's no licence issue here because I'm not distributing the result. GPLv2 is awesome in that respect.
 3. It's a bit of a liberty for my 14kth post but what are you going to do about it? In response to the comment about displaying it as 14k, I have no problem with that. When you're < 1000 posts, each post matters. When you get to the thousands, nothing nearer than the hundreds mark even matters. 13.9k to 14k matters. 13k to 13.1k is an indication of progress, in a way that 13099 to 13100 isn't. I'd be quite happy to see post counts formatted that way in main display.
 4. Especially if your figures happen to be: Most Online Today: 1,555. Most Online Ever: 3,726 (January 18, 2008, 10:00:17 PM)
Other software / SMF with a different name, huh?
« on June 7th, 2013, 05:44 PM »
I mean Wedge, a fork of SMF is basically SMF with a different name minus a few minor features from what I have seen.
-- a former beta tester, yesterday. So, let's put the record straight. A year ago, I compiled this list. Let's see what's in there now. Interestingly it's been a year to the day since that previous post.

Auto embedding
Posting and PM drafts
Flexible skin support using LESS syntax, including vendor prefix abstraction (and changed default theme, of course)
bbc/WYSIWYG editor in the quick reply area
Thoughts system
Per topic feeds
Bad Behaviour integrated by default
IPv6 support
Template macros such as <we:cat>
JavaScript and CSS files are minified by default, and JS files are automatically deferred to late in the script for extra smoothness
Merge double posts
User and action menus in posts
Automatic quote splitting
Spoiler, footnotes, more bbc
Gravatar support
Using a board as a blog
A revised plugin manager that supports web based uploading WITHOUT CHANGING FILE PERMISSIONS AT ALL
Moderation filter system that allows admins to define rules for content management (e.g. moderation on a post automatically for containing rude words)
Q&A that supports multiple languages and multiple answers per question
Responsive sidebar
Displaying of multiple badges per user in the poster information and drag 'n' drop to configure what order they are shown in
Ability to use colour, bold, italic, underline and arbitrary CSS for all user names based on their primary group
Notifications system
Ability to edit registration agreement from the admin panel without requiring file write access, and also forcing users to re-accept it (and either locking them out the forum or simply barring them from posting until they accept it)
Ability to search and edit all language files, including plugins, from the admin panel, including showing both the master and current values
Ability to edit email templates from the admin panel
Ability to see who voted in a poll with proper privacy controls
Display reported posts in the user's profile
Rewritten ban system, including IPv6 support, ability to handle GMail type emails (, = username and similar constructions)
Rewritten warning system, including selective punishments per warning including but not limited to removal of signature, removal of avatar, posts being altered, moderation, bans
Ability to 'soft ban' a user, similar to Annoy User
Ability to send individual items out of the mail queue
Mass move topics is now more selective than just 'everything in a board', you can pick boards to move from, as well as other criteria
Merge two user accounts together
Moderation centre shown on the board index as if it were a real category (but it isn't a real category, it just looks like one)
Centralised popup system even for general prompts and OK/cancel notifications, which is styleable (unlike standard browser ones)
Custom fields now have proper privacy controls, are drag 'n' drop to rearrange the order and can optionally be shown on the memberlist too
Pinned topics can be drag 'n' drop rearranged
Group access can not only be given to boards, it can also be revoked (unlike SMF)
Board access is also split between seeing it and entering it, including the option to add a custom message to users who can see a board but not access its content
The SMF 'member options' area is replaced with one that doesn't have 'per theme' hassle and is actually quicker to use to set what you want
Proper timezone support for users (set timezone on registration, not a number of hours offset... even daylight savings works perfectly)
News items are also drag 'n' drop for reordering, show the bbc editor interface and also have privacy options attached now
Better internationalisation support in general
Logins can be username/email, username only, email only
Customised selectbox that shows images and other HTML (useful for selecting icons)
Search removed from the main menu and made in to a sexy popup
Use of HTML5 inputs wherever possible
phpinfo() shown in the admin panel
Your buddies/contact list is shown in the PM area with a nice clicky interface
Admins can now set which boards can be ignored by users
When splitting a topic, the new topic can be in a different board
When issuing a redirection, you can also set the redirection topic to actually redirect you automatically (and if it's to a board you can't see, a nicer error is shown to you)
Topic privacy
Topics have a good meta description now

This doesn't include the various and numerous, SMF bug fixes or the various underlying changes required to make any of it work.

Currently being added
Infinite scroll in topics
Proper Sphinx and ElasticSearch support

Default theme
Admin front page
Completely new CAPTCHA even down to including a new font (and replacing all the old SMF fonts)
Better page navigation
Revamped theme layer system
Some code rewritten to use better OOP
PMs automatically save to outbox, there's no option for it, nor is it something you can disable
Improved newsletter interface

Non UTF-8 support
PHP below 5.2
move, glow, shadow bbc
SMF's buggy package manager
Support for really bad hosts like 000webhost
Spell checker
Fulltext searching (custom indexes are almost always better, if Sphinx/ES aren't an option)
"Simple" permissions
Calendar - now in a plugin

There are probably other things I've forgotten, too. But of course, it's just SMF with a few new features, right? RIGHT?!?! And it's not like we're adding new features or anything.

Of course it's not. I don't know why I'm letting myself get annoyed. I find it amusing that the person who makes such comments 1) does so in a place where he KNOWS I won't normally see them and can't defend myself or Wedge (someone considerately informed me via PM), 2) came here only a few days ago to ask for my help with his project, after all the shite I had with him a bit back and seemed *surprised* that I wasn't interested in helping them and 3) tells me off for things before doing them himself

Mind you, said person is talking about his own forum software in Python and Django which will obviously be better than anything else out there. Assuming he ever finishes any of it. This is after both his own PHP based software, and his subsequent SMF fork, have already fallen by the wayside.