So, it turns out I was wrong about Microsoft's IE8 standards-mode rendering switch - they're proposing that web authors use either a meta element or an HTTP header as an opt-in.
My initial gut reaction: wow, that sucks.( cut because it's web-geeky and looong )
So: flip the default around, and this is a good idea. Otherwise, it sucks.
Finally, a word from the lemurs.
- Put your music player on random.
- Find photos of the first 20 artists/bands that come up (no repeats and no cheating).
- Have people guess who the artists/bands are. (comments will be screened until later)
- Make people do the same on their journal. If they feel like it.
- Answers in due course.
A few weeks back, via the Web Standards Project, I learned of the Email Standards Project, a similar effort to encourage the adoption of HTML/CSS standards in email clients when rendering HTML emails.
All I can say is: about bloody time. Getting HTML mails to render consistently across different mail clients is a monumental pain in the arse. There are mail clients out there with HTML rendering in the same league as Netscape 4.x (translation: horrible) and it ain't just small fry, either. The ESP people highlight Google Gmail, Lotus Notes 8 and Outlook 2007 as among the worst offenders.
I'm disappointed, but not surprised, to see the announcement immediately attracted the usual rants about how HTML mail is evil and should never be sent/encouraged/tolerated/etc. I've encountered this viewpoint many times before, and while I'm sympathetic to most of the technical arguments1, I have to say: get over it, people.
To get rid of HTML email at this point in proceedings would require removing functionality (in some cases, default behaviour) from the top 10 or so email clients (yeah, Microsoft are soooo likely to do that2) and significantly altering the behaviour of 99.9% of the email-using population of the world. You'd probably need to kill every marketing person in the world too. People like HTML mail. They like to be able to use bigger text, fancy fonts, and even (shudder) marquee and blink elements.
Ranting against HTML mail is an exercise in futility. The horse hasn't just bolted, it's sired offspring from every mare in the country, taking time out half-way to come back and nick the mints out of your pocket. Trying to shut the stable door at this stage just makes you look like a foolish, dogmatic, elitist zealot. This is the sort of thing that leads people to view standardistas/open source advocates/linux-weenies as a group of wild-eyed lunatics who should be avoided like the plague, and it's counter-productive, because it drives ordinary users away from even contemplating a better way of doing things.3
The ESP is an important part of a far more productive approach: make HTML mail work better for everyone. Speak to software vendors, and get mail programs to respect user wishes better, so that those who want to sent plain text mail can do so easily. Get mail readers to render HTML mail in a sane way (this, in itself, stands a good chance of significantly reducing the badnwidth usage - a lot of the current bulk is because people have to use 1997-vintage table layouts and tag soup to make it look sensible in major mail clients). Get mail senders to include plain text alternatives and build the mail properly so that people who read text-only can still read them. The last couple add up to improve accessibility, too.
1saps bandwith, harder to construct programmatically, etc. There are compatibility arguments re: older mail clients, but most of them involve badly constructed mails which don't include text alternatives (see! they're trying to save bandwidth by not duplicating stuff!) or mails which have fallen foul of the appalling HTML rendering of some mail clients.
2Though they did completely shag the HTML rendering in the last outlook release - previous versions of Outlook were much better than Outlook 2007.
3To borrow liberally from a stock response to spam "solutions":
Your approach advocates a: (x) technical ( ) legislative ( ) market-based (x) vigilante approach to making email better. Your idea will not work. Here is why it won't work. (One or more of the following may apply to your particular idea, and it may have other flaws too.) (x) Users of email will not put up with it (x) Microsoft will not put up with it (x) Most other software vendors will not put up with it (x) Requires total cooperation from all email software vendors at the same time (x) Requires total cooperation from everybody in the world (x) Many email users cannot afford to alienate potential employers/customers Specifically, your plan fails to account for (x) Lack of centrally controlling authority for email (x) Popularity of presentational twiddly bits like colours and fonts (x) Public reluctance to accept removal of existing functionality (x) Huge existing software investment in HTML Email (x) Outlook (x) The entire marketing industry
(and probably a whole bunch of other reasons I left out)
After a long silence, Microsoft have finally started talking about IE8 features, and it's actually good news - IE8 internal builds are now passing the Acid 2 CSS test1.
While many will hurl (not entirely undeserved) general derision in
MS's direction for being the last of the major browser vendors to get
there2, the important point is they got there. The
Acid 2 Test is eeeevil. This means IE8 has substantial and wide-ranging
fixes to their CSS support (including, but not limited to, position:
fixed, float/clear, margins, generated content, ignoring bad
declarations, display: table and associated gubbins) and also fixes to
more obscure bits of HTML like the
IE7 had quite a few fixes to CSS support, and was welcomed for that reason, but still lagged a bit behind the competition. If IE8 is passing acid 2, that's a huge leap forward. Once this version hits lots of users (probably a good 2-3 years after they release it, at least) this will finally open up to general use several areas of CSS2 that are currently off-limits. Headlines: position: fixed (menus that stay fixed in the window while the page scrolls, without frames), display: table (table-style layouts without table markup), generated content (tricky to describe, but it allows all sorts of cunning stuff). It will also significantly reduce the pain of making float/clear work cross-browser.
1Unfortunately, someone managed to actually break the main Acid 2 test site recently, but the powers that be are on the case, and it should start working again sometime soon.
2Out of the big four, Safari were first (internal build 27 Apr 2005, general release (v2.02) 31 Oct 2005) then Opera (public experimental build 10 Mar 2006, general (v9.0) 20 Jun 2006) then Firefox (semi-public dev build 12 Apr 2006, general release (v3.0) early 2008)
Superficially pretty, with the high detail models, subtle distance haze, lighting effects, and so on. But to my eye, once you get beyond extreme close-up at high resolution, all those details seem to smoosh together into a sort of grey blurry mess. This looked horrible at the 800x600 resolution the game defaulted to; it's better, but still an issue, at 1280x1024. There's lots of fine detail, but when that disappears (i.e. more than ten yards away) there's no intermediate-level detail to replace it.
Deathmatch, Team deathmatch, yawn. Been there, done that, having waaay too much fun with TF2 to care any more. My tastes in online shooty-goodness aside, it seems generally well done. Movement is slick, map design seems good (the two in the demo are a bit less claustrophobic than many UT2k4 maps, which is a plus for me) nice range of weapons, with a mix of old favourites from past Unreal games.
I'd be far more interested in the "warfare" mode (which is sorta "onslaught-with-knobs-on", I believe; Onslaught was my main reason for playing UT2k4) but I can't seem to find it in the demo.
I never played Vehicle CTF in UT2k4, but I fired it up to have a play with the vehicles, in the absence of Onslaught/Warfare. They all seem kinda slow compared to the old ones, which is unfortunate - some of the best fun to be had in UT2k4 was blasting around the landscape in a fast car, squishing enemies and performing ludicrous stunts.
The scorpion's new gun is more conventional, possibly more effective in many circumstances, but also more BORING than the old energy-ribbon-thing.
I'm not sure that turning the Hellbender into a two-seater and giving the driver control of the skymine is a good idea (the old way encouraged teamwork, as a 1-man HB was vulnerable, but a fully-manned one was downright dangerous to lighter vehicles and infantry). The new visuals of the skymine projectiles, while very pretty, make it damn difficult to see what you're shooting at, particularly when you combo (which is pretty much compulsory if you want to kill anything).
The raptor doesn't feel as fast and maneuverable as before, and I miss the roar of the old secondary-fire missile as it arcs away on a smoke trail - the new one is downright dull by comparison.
General Design Stuff
I've got lots of little niggles here.
In UT3's team deathmatch mode, player avatars are tinted their team colour. I think this has been done because it's the only way to get any sort of sane IFF, given the incredibly detailed/varied models flying around (and the "grey blurry thing" effect). Problem is, in amongst all the highly-detailed, realistically-lit models, it looks kinda silly. Part of what bugs me here is that UT2K4 did a pretty good job of this. As soon as you went team-based (rather than free-for-all) all the models gained huge slabs of primary team colour, and little coloured icons over their heads.
The game menus seem strangely unresponsive. There's an appreciable pause between mouse click and something happening. The menu items themselves are teeny-weeny things, in the middle of acres of space (OK, there's a fisheye zoom on the menu, so that items get bigger as you mouse-over, but that just makes them moving targets). This is a minor niggle, I know, but it's just amateur to have the first screens the user interacts with be such a pain to use. Again, UT2k4 did it better, if less prettily.
As for the very first welcome screen being "username and password, please" (which still often pesters me even after I've ticked the "remember password" and "login automatically" boxes, and frequently fails to login altogether)... Grrr.
OK, this is a pre-release demo, and some things may well be fixed before the final release, but I doubt it's all going to magically get better.
- The engine is an amazing technical acheivement, in terms of the amount of detail it can throw around, at high resolution, with decent framerates.
- From an artistic standpoint, the models and textures are fabulous - up close. Unfortunately, at anything beyond in-yer-face-claustrophobic-deathmatch range (and the DM maps are more open than before) it's all very same-y.
- The game they've built on all this? Not so much. I'm not finding any positive changes to the gameplay compared to previous versions; I'm not impressed by the new vehicles, and from a servicing-the-gameplay standpoint, the visuals are actually a step backwards. It feels like the "fun" element has been lost in the quest for stunning high-res screenshots and a high-powered engine.
- I've been spoiled by TF2, in multiple senses. Not only have Valve provided such a high level of game design that other games seem boring by comparison, but the commentaries and interviews allow me to analyse and articulate the reasons WHY the other games feel dull. Scary.
TF2 has reminded me just how much FUN games can be, and the UT3 demo just doesn't hit the spot. Which is a shame, as I was really looking forward to it. I'm still interested in "Warfare" mode, but I'm not really going to buy the whole game knowing that all the game modes I have played are not of interest, just to check out one I haven't.
So, it's time for Gav's Fortress Forever (FF) versus Team Fortress 2 (TF2) showdown post.
First, a bit of background. Before Counter-Strike became the most popular PC team-based online shooter on the planet, that crown was held by Team Fortress Classic, a Half-Life mod which was itself a remake of an old Quake Mod named Team Fortress. TF/TFC is class-based, each class having strengths and weaknesses (the Heavy Weapons Guy is slow but well armoured with a big gun; the Scout is fast, but lacks armour and firepower). The pace of TFC is slower than deathmatch games, and therefore less "twitch"-reflex based, with a bit more emphasis on being cunning.
Back in the day, I played TFC a lot, and only gave up when the online community withered to the point where it was almost impossible to find a fun game. So I'm definitely in the market for a follow-up (well, I was, because I've now bought/downloaded both of the main contenders).( cut for the sanity of those not interested in online multiplayer shooter games )
To summarise the two games: FF is effectively TFC with the annoyances fixed. TF2 is a much more radical overhaul of the whole Team Fortress concept.
If you're an old TFC hand, appalled at what Valve have done with TF2 (no grenades!? teleporters?! minimal CTF!?) then go check out FF. Ditto if you're an old-school FPS player who finds TF2 too slow-paced.
If you're generally intimidated by online multiplayer shooters because the twitch-reflex gamers hand you your ass on a plate 2 seconds after you spawn, try TF2. It's a slower pace, and most classes reward tactical cunning more than twitch reflexes
If you're new to this whole Team Fortress lark, then I recommend you try both. While there are many similarities, they're rather different games, and different people will prefer different flavours.
My preference is for TF2. This is mostly due to the "fun" factor, but I think it's also down to the TF2 developers evolving the game in pretty much the direction I wanted it to go.
I really feel for the FF developers. They've worked hard for the last couple of years producing a really high quality multiplayer mod, which makes numerous improvements to the TFC game style yet keeps the overall feel, only to release it right in front of the oncoming TF2 juggernaut. I hope they don't get squished, it's a damn good mod.
So, having waffled about Fortress Forever last week, now it's the turn of Team Fortress 2. Again, first impressions are very good.
- It looks and sounds fantastic. Think crazy-60s-spy-movie-with-mad-scientists-
kitsch, via The Incredibles.
- Gameplay has been streamlined considerably; no grenades (apart from the demoman's grenade launcher) and most classes are down to one each of main, backup and melee weapons.
- In the process, that means conc-jumping and bunny hopping are out (I regard this as a good thing)
- The pace is, if anything, a little SLOWER than TFC.
- much less Capture-The-Flag. Out of the 6 maps in the release, only one is CTF - a remake of the venerable 2fort. In its place is Warpath-style control-point action (including a radical reworking of Well) and Dustbowl-style attack/defend (including an update of Dustbowl).
- The most-changed class is almost certainly the medic, who is now actually likely to run around healing team-mates, rather than acting as a lone-wolf, leet-skillz-powered, conc-jumping offensive class.
In short, it's a big change from TFC. Fortunately, it has the gameplay to back up the looks - it's just buckets of fun, with a serious side order of "just one more round".
It's also funny. Voice acting and scripting is superb. It's the little touches that make it - when a HWG really lets rip with his assault cannon, he'll start laughing maniacally (with matching visuals); everything the Pyro says is muffled to "mmmph!" by his gasmask, and so on.
Seriously, it's almost completely unrecognisable. After playing it a couple of times, I finally started to notice the similarities, and I knew the old one inside-out and backwards.
As some of you may remember, I was a big fan of Team Fortress Classic, back in the day, but gave up playing a few years ago. Well, us Team Fortress fans are being spoiled rotten this week. The official Team Fortress 2 goes into public beta in a few days, but just beating that out of the gate last Thursday was the first release of Fortress Forever (if you don't want to wrestle with the horrors of game site download arrangements like FilePlanet, there are official torrent links here)
I've not had time to form a detailed opinion yet (I only found out about the release yesterday afternoon, due to heavy-weapons-lurgy leaving me completely incapable for most of Thursday) but here are some first impressions:
- As mods go, it's pretty slickly put together; masses of generally high-quality new models, textures and sounds, even a proper windows installer, to avoid all that yucky messing about with zip files
- They've made a solid effort to put hints and tips in place, to help newbies figure out the complexities of TF
- Based on a quick runaround, where they've remade TFC maps, most of the changes from the original seem to fix annoyances (for example, the centre bridge area of Crossover2, always a horrendous choke point, has been completely redesigned to avoid it being a total grenade-filled traffic jam like it used to be). In some cases, they've made apparently-subtle geometry changes that I suspect will actually have a radical effect on the gameplay dynamics of the map; I won't know for certain until I've played more of them.
- They fixed the Pyro flamethrower to the point where it's actually useful again. hehe.
- They appear to have crippled the Heavy, with an overly-onerous "overheating gun-barrels" thing, but there's stuff in the forums from the developers saying that's going to get changed pretty drastically in the near future.
- The pace is generally a little faster than TFC
- They've added a primary scoring system, always visible in the HUD, which includes lots of points for non-frag-related activities - so medics get points for healing team-mates, engies get points for fixing SGs that aren't theirs, and such things. Not sure of all the details yet, but it seems good.
- initial gameplay impressions are really good - my first game was on "hunted", an old favourite of mine that became pretty unplayable in the latter days of TFC due to lack of players; it played really well.
I'm still a little worried that it may have suffered from some degree of Grognard Capture (Excessive nerfing of the HWG; I think the speed increase will favour the leet-of-reflex more than the sharp-of-mind; TFC's reversing of that balance, compared to most FPS games, was a large part of the appeal, for me at least) but only time and gameplay will tell on that front, and first impressions are good.
Overall, it looks pretty good. Much kudos to the FF team on a very solid first release.
 It boggles me that more games, both amateur and pro, don't make torrents available; this goes double for demos of pro games; the demo of a highly anticipated game will have masses of people wanting to download it in a short space of time, and probably be pretty big (1GB+ is not that unusual). So why do they persist in restricting distribution to blecherous download "services" like FilePlanet, rather than using torrents, which are designed to handle this precise situation?
 Rough timing: 2 seconds of continuous fire (a large chunk of which is restricted by the wind-up time) followed by 1 second of forced cool-down; combined with the slightly increased rate-of-fire of the soldier's RPG, this seriously re-jigs things against the HWG. I agreed with the veteran's consensus towards the end of the TFC era that the HWG needed to be weakened a little, but this is way too much. Have to wait and see what the developers' promised revamp does...
The works of Douglas Hill - primarily the above-mentioned Last Legionary series, but also the Huntsman and Colsec series - were probably my favourite SF before I graduated onto the adult stuff. They may not have scaled the heights of literary achievement, but I loved 'em when I was 9/10/11. He was one of a small band of childrens' SF writers who introduced me to the idea that I might be able to get some of my spaceships-and-rayguns-and-aliens-and-
Without that shove to cross the media boundaries, and given the (lack of) quantity/quality of much kiddie-friendly early-mid 1980s TV/Film SF, it's not too much of a stretch to a parallel world where my interest in the genre died in 1983 and I never met most of you lot, including calatrice.
I was fortunate enough to meet him at Eboracon (Unicon 2001, organised by psycho_machia and others, IIRC) and he was a very normal, friendly guy, who seemed to mostly be interested in writing stuff that got kids reading, and literary fame could go hang. Worked on me.
There's a new RTS game out, called Supreme Commander. It is, to all intents and purposes, a direct sequel to Total Annihilation (but they're not allowed to say so for legal reasons)
Why do I think this is special? The traditional RTS control system, which pretty much dates back to Command & Conquer and Warcraft 2, becomes a major pain in the proverbial when confronted with large numbers of units (say, greater than 50).
Most RTS games have countered this by coming up with various mechanisms to limit the number of units in-game. TA, on the other hand, changed other things to make it possible to cope with large armies (unlimited build queues, unlimited resources, big maps, big guns, relatively smart units).
Supreme Commander continues that evolution. Most of the innovation is in the UI - the level of control of building and factory production makes every other RTS I've played feel painful. You can assign one factory to assist another, so they co-operatively chew through the same build queue; there's a "repeat this build queue" button so you can leave factories producing your army while you go do something else. Also, the BEST mechanism for handling transport units I have ever seen, which extends to getting transports to assist factories, so that units come straight off the production line and load onto a transport, which then transports them to the rally point halfway across the map, without you having to lift a finger. It's actually possible to get your 200 tanks across a large expanse of water without getting RSI or dying of boredom.
Plus, it's great fun to march a group of MonkeyLords (warning: dialup-unfriendly picture) into your enemy's base and watch them chew it up.
It's not all rosy - some of the pathfinding can be a bit bizzare, especially when attempting to formation-move large naval groups, and experimental units can sometimes get a bit confused and bimble about aimlessly for several seconds, rather then fulfilling their orders (which are typically something like "I'd like a half-kilometre-wide swathe of destruction through the enemy base, please"). Also, SC is a bit of a resource hog. It really wants to be played at a resolution of 1280x1024, at least, if you want to see what's going on properly. I've got an Athlon 64 X2 4600+, with 2GB of ram and a Geforce 7900 GT - not quite an uber-leet gaming rig by today's standards, but it's no slouch, either - and SC has made it stagger under the load a few times.
Overall, though, it's a more-than-worthy successor to the mighty Total Annihilation, and the game currently occupying most of my (rather scarce) free time.
If you want to check it out, there's a bandwidth-busting (1079MB) demo available (which is also on a lot of game magazine cover disks at the moment).
Via slashdot, I learn of the passing of John Backus (NYT obituary) co-creator of the Backus-Naur Form, and the guy who led the team that developed the Fortran programming language. Predictably, practically the first comment on the slashdot thread is knocking Fortran.
Back when I was studying engineering at University in the early nineties, I had to learn Fortran - my final year project involved writing a moderately substantial program using it. Well, the biggest program I'd ever written at the time. I learned a lot about the day-to-day stuff of programming from that project; it was probably the formative experience that convinced me that I'd quite like to do such things for a living. As a result, at times like this, I feel a need to leap to the defence of the language.
FORTRAN (from "FORmula TRANslation") may have flaws, but comparing it to C is like comparing the Wright Flyer to a Sopwith Camel. Yes, the Camel is the better plane, but I don't hear many people complaining about the Wright Flyer's poor rate of climb - as the name implies, it was pretty radical that it flew. Fortran, developed in the mid-to-late 1950s, was the first high-level programming language to see widespread use. Before it came along, programming was done in assembly language, if you were lucky.
In this context, "high-level" means two main things: "approximately readable by normal people" and "you don't have to worry about the precise design of the CPU you're running on". Assembler language is neither - it's pretty much a one-to-one mapping of the raw machine code into easier-to-remember mnemnoics like "LDA", "JSR" or "BNE" (I said easier, not easy), you need to keep tabs on exactly what data is currently in which bit of the CPU and tell the cpu when to ship stuff out into the main memory.
High-level languages like Fortran separate the programmer from most of that stuff by providing a more human-friendly language (usually built around a combination of real words and algebraic-style expressions) for the programmers to work with, and a "compiler" that translates the human-code into the machine-code. The designers of Fortran made a few mistakes, which have been fixed in many newer languages, but they didn't exactly have the opportunity to learn from the mistakes of others, because no one else had done it before. Criticising Fortran for its lack of (for example) Polymorphism is like complaining that the Wright Flyer didn't have seat-back DVD players (the early models didn't even have SEATS).
For all its faults, Fortran (albeit in an evolved form) is still in widespread use today. As its expanded name implies, Fortran is not a general-purpose language - it was invented specifically for the purpose of mathematical computation. They added native support for Complex Numbers well before support for characters (i.e. text) and there are some extremely high-powered mathematics code libraries available. If you want to do some seriously heavy-duty computational fluid dynamics, Fortran's still a contender. Which is why they taught it to me, 'cause that's the sort of thing aero engineers do.
Not bad for a programming language that's over half a century old (a milestone that the grand old man of programming languages, C, won't reach for another 15 years or so).
So, RIP John Backus. Thanks for inventing the most important tool of my profession.
For those who are interested, I posted a couple more pics on my journal this morning; I twiddled the bit that prevents it showing up on the friends list, as I figured that most people have limited interest in baby piccies.
I've been very lax in recent months about writing these things up. So, a brief run-down of some recent viewings:
Bloody Brilliant. I've not read the book (and given calatrice's enthusiastic anti-recommendation, probably never will) but the film is powerful, moving stuff. A proper grown-up intelligent British1 SF movie. Go see it.
Warning for the squeamish: it's also kinda violent; OTOH, Cal (who is normally squicked by anything more bloody than Toy Story) seems to have coped fairly well.
1OK, the writer/director's mexican, and I've no idea who paid for it, but it's set in blighty, was mostly filmed here, and *feels* British, so that'll do me.
Battlestar Galactica 3.1-3.4
Also brilliant, for several of the same reasons. I don't think I can say much more without spraying spoilers everywhere.
Not bad. It's precisely the sort of Saturday tea-time swashbuckler they don't make any more. My 10-year-old self would have been glued to it every week, despite the rather frugal production values and spotty script (Contractions, guys? You don't have the excuse of being androids). However, my 34-year-old self is having an apathy attack and probably isn't going to bother watching any more of it.