T=Machine

Almost a year after Apple’s disastrous “force consumers to download Lion, instead of installing from DVD”, apparently it still doesn’t work. It’s hard to recommend OS X to anyone after this experience.

UPDATE 2: Apple’s “download a file from the internet” code is so bad it’s causing the MacBook to overheat – 80 degrees celsius, very close to the “automatically reboot” temperature. This is *to download a file*. Apple’s misuse / misunderstanding of web technologies seems quite incredible.

(the process is called “storeagent”)

My last 24 hours:

1. Buy Lion
2. Download starts
3. …it’s a 4gb download, this takes a long time…
4. Download stops at 25% for no reason.
5. Resume button gives a wait cursor for 5 seconds, then goes back to “paused”
6. Repeat twice
7. Third time, the Resume button is disabled, and now Lion is stuck in “Waiting” and there’s no buttons you can press except “cancel”
8. …
9. Remains in “waiting” for many hours. Googling suggests this is a permanent crash in Apple’s App Store.
10. Cancel the download, re click the “buy app” link
11. Apple quits OS X, kills all apps, deletes all unsaved data, throws me out to the login screen
12. Login again, and Lion icon has appeared in the dock.
13. …but: Lion now refuses to even start downloading – it’s stuck on “Paused, 0 of 0 bytes”

UPDATE:

1. Try again (delete OS X Lion, re-purchase from App Store) and … finally the download starts. Waiting now to see if it will complete this time, instead of giving up partway like before…

I.e. Apple’s infrastructure is still blocking me from downloading the OS. How hard can it be to *download a file* ?

Next step: walk in to an apple store and ask them to give me a USB stick, since their webserver is FUBAR.

Filed under : Web 0.1, bitching, web 2.0

2 Comments »

On Facebook.com front page:

“Dave Stone and Mike Merren recently read articles.”

…so I click on one of them. Lo and behold, instead of getting an article, I get The Independent app trying to force me to install it.

Force. Yes, force – there is no option to “view the article”, the only option is “install this app or cancel”.

Lots of people complaining about this recently, but there’s really only two things to do:

1. Shame the newspaper by tweeting them
2. Report them to Facebook for abuse

Tweet them to let them know what you think

@Guardian and @TheIndyNews

Report to Facebook

Yes, I know Facebook is behind this in the first place, but if you don’t bug them about it, they’re less likely to care:

…and you might as well let the developers know how pissed you are – directly – since Facebook gives you this option:

Filed under : bitching, facebook

0 Comment »

StackOverflow.com has long had one of the worst search-engines I’ve ever seen. It’s clearly a simple thing hacked together. It generally doesn’t work, and most of the people I know use google isntead, and rely upon Google to collage all the stackoverflow results together.

Occasionally, you have search terms where Google gives you lots of non-programming hits (e.g. “iphone video (something)”. So the above method fails, and you have to use the appalling SO search engine.

Then you get this, because the search engine is so poor that it often ignores search-terms, so you have to creatively re-search and experiment to find the results you need:

ARRGGGH!!!

Filed under : bitching, community, web 2.0

10 Comments »

…as in: after 5 odd years, on OS X the official uploader still “requires” you to either lose all your data every time it stumbles, or … force-crash it. Which, paradoxically, keeps your data intact. Confused? You should be.

e.g. you get 50% through uploading a few hundred photos, and your broadband has a momentary slowdown. Ten seconds appears to be all it takes. Because the flickr app doesn’t do basic error handling, it’ll hang at this point – forever.

If you do the obvious thing and hit “cancel” (there’s no “retry” button – why would you want to retry?), it deletes your data.

If you quit, it also deletes your data. (this is the mistake I made just now. That’s 20 minutes of editing image data I now have to do all over again. Sigh)

The only options are:

1. pull out the network cable, causing it to hard-crash … and “enable” the retry button
2. force-quit the app, causing it to crash … and when you restart it, it will automatically load in all the data

So, note to self: if flickr uploader hangs, FORCE KILL the ****er. Don’t do anything sensible or sane – it won’t work.

And … note to flickr: there’s quite a lot of Mac users these days; might be a good idea to start supporting them.

Filed under : Web 0.1, bitching

1 Comment »

Why is it that these days every time a London-based “digital agency” or “mobile agency” gets hold of your email address, they IMMEDIATELY sign you up to their spam mailing list? Even some outside London have started doing this too.

YuzaMobile is the most recent example – why do it? It doesn’t benefit them: their spam has just thoroughly convinced me that I never want to let any of my clients or partners anywhere near them. They have a cavalier disregard for basic comms etiquette.

How did they get my email?

Well, I sent a single personal email to one of their directors who I’d met at an event.

How does that square with “spam me now, please!”?

I have no idea.

Filed under : bitching, marketing and PR

4 Comments »

I just waited 4.5 hours for Photoshop Elements to install.

And what happens?

It crawls to 100%, Adobe finishes the process, then announces:

I didn’t install Photoshop Elements. I won’t tell you why. [So you can't do anything about it]. But [if you're really stupid, and] you want to try again, run the uninstall program first, then start again.

…but it *did* install the *one thing* I told it not to: the infamous Adobe viruses.

(and, of course, it installed the UNinstaller … even though it didn’t bother with the main app)

This is why, when Adobe finally goes bankrupt, or is acquired and dissolved into nothingness, I shall be cracking open the champagne and cheering their demise :).

EDIT: incidentally, it’s also why I’m now looking for a pirated copy. I’ve got the legitimate copy sitting in my hands, I’m legal, but Adobe’s sheer incompetence means it refuses to install. I’m pretty sure a hacked copy will “Just Work ™”, and contain fewer viruses.

Filed under : bitching

0 Comment »

Is there a place to complain that UK government departments are breaking the internet standards and refuse to fix their websites?

Occasionally, you find sites that do this. Usually, when you tell the organization, they’re a little embarassed, and rush to fix them.

From HMRC, I got a polite, pedantic, *but entirely incorrect* response telling me that the “standard” was X, when I know that to be false (as does anyone who has read the offiicial standards, as documented by the Internet RFCs).

They apparently can’t be bothered to read the standards, and don’t care that they’re wrong.

No wonder so many people hate civil servants: holier-than-thou attitude coupled with being clearly, inarguably, wrong. Sigh.

Filed under : bitching, community, web 2.0

0 Comment »

Some people are telling me I should just let it go, but honestly I just can’t do that. I’d rather quit.

(As they say: It’s funny, becaus it’s true. If you develop for Apple platforms, that eloquently sums up how Apple (currently) appears to everyone they partner with: a childish, passive-aggressive approach to everything)

Filed under : amusing, bitching

2 Comments »

(…or else forgo it)

(EDIT: To be clear: Piracy isn’t theft, but it certainly is illegal. Please do not misconstrue: I do not condone piracy; this post is a lament at the extent to which the retail industries encourage or coerce consumers to pirate content. I am still looking for a legal way to buy the digital data I want, and in the meantime, I have Spotify…)

I want a single that came out 5 years ago. It’s available to purchase on iTunes …. in the USA.

I’m “not allowed” to give Apple money to buy that track, because my account was originally created when I was sitting in the UK. IIRC, even when I’m physically in the USA next month, I will still “not be allowed” to give them money for this (but … who knows? Apple doesn’t bother explaining this stuff to the normal consumer)

Switch to UK iTunes “mode”, and … Apple does not sell that track in the UK.

So, once again, the music industry would prefer that I go and rip the MP3 than that I *give them money*.

Do they care? Do they even know?

Of course not.

They will *never know* that I did this. They have no mechanism to allow me to *tell* them that I attempted a purchase – and was rebuffed. This would cost them nothing, but … they can’t be bothered.

Equally, when I rip the MP3, they’ll never know that I did. It has literally zero effect on their business. Because piracy is not theft: digital data is not physical property, and copying does not affect the original in any way.

Sigh. One day, the digital industries will grow up. I hope I’m still alive to see it.

Filed under : bitching, web 2.0

1 Comment »

Let’s get this straight: if we judge him solely by output (games), then he is not a genius – he’s lazy. Everyone knows the 1% inspiration/99% perspiration quote, and – looking at the last shipped title – IMHO it’s inexcusable to ship crap and pretend it’s OK. You can’t just abrogate responsibility once you stick your name on Spore…

(disclaimer: when I say “lazy” I don’t mean universally; I mean that in at least one crucial aspect, he failed to apply simple due diligence to his own named project; arguably, it’s a kind of laziness in itself not to have checked this stuff, or a kind of cowardice not to have insisted it be done “correctly”; but this post is really about the overall impact of the game, and the way that an individual, if they were to stamp their persona on a project – and expect us to read their persona from interacting with the product – comes across. I have no idea what Will Wright is like as an individual; this is a post about Will Wright as the PR entity…)
(more…)

Filed under : bitching, computer games, games design

6 Comments »

I’m trying to download the 3.0 OS update for iPhone…and being denied by Apple’s own software – that cannot even download a single file from a website (!)

It’s a 1GB download that you “must” download via iTunes, because … well … because … um … Apple hates web browsers? I don’t know. Hard to see why it is downloaded via iT at all, really. It is rather strange.

(EDIT: it has now dropped to being a 230 MB download; I have no idea why – it was only a hundred meg or so into the alleged 1 GB download when it crashed, and when I retried it became 230 MB. Odd…)

And yes – it really *is* downloading a website file (that’s all it’s doing):

GET /content.info.apple.com/iPhone/stuff.stuff/iPhone1,2_3.0_7A341_Restore.ipsw HTTP/1.1
Host: appldnld.apple.com.edgesuite.net
User-Agent: iTunes/8.2 (Macintosh; N; Intel)
Connection: close

That’s missing a key line. The line that resumes the download from where it left off. Apple apparently decided to write a “crap” web-browser, and embed it inside iTunes. Why? Why, when they have one of the world’s best web-browsers, do they insist on writing an extra one – and missing out fundamental basic features (like resumable downloads)?

There are occasional latency spikes on my net connection. iTunes is such a terrible “web browser” that when this happens, it arbitrarily (note: no other web browser would do this!) decides to cancel the download. There is no “resume” option and no “retry” option.

Congratulations, Apple! Having 2 copies of the same “core” software, one which works and one which doesn’t, and not allowing the user to use the “good” one when they need to? You’re well on your way to becoming Microsoft :).

Filed under : Web 0.1, amusing, bitching, iphone

4 Comments »

Over the past few weeks, I’ve been looking at a lot of independent developers’ websites. It’s quite surprising how many of them go out of their way to make their site unusable – clearly thinking that they’re achieving the opposite. But also, today, Wikipedia started actively doing a very minor (but no less irritating) content-block on mobile users. And last week, I found one of the main games-news sites is also actively *hard*-blocking mobile users.

This was annoying (and stupid!) 5 years ago, when sites added the “smartphones” to their content-blocking, even though smartphones could (and happily would) render full-fat webpages perfectly (tabbed browsing worked fine in Opera on Windows Mobile back in 2005 – I used it a lot).

Now, with the iPhone added to the list of clients that these sites are blocking, it’s a bit worse: Apple won’t allow you to purchase any web browser other than their version of Safari, and Safari won’t allow you to lie to the website and tell it you’re not using a cell phone (this was the standard workaround on windows mobile/opera for stupid web design teams: tell Opera to claim your cell phone was a Windows desktop). The iPhone, with a better quality web-browser than many desktops currently run? That’s just insane…

Wikipedia: mobile users, go away

Until/unless they decide to fix it, it’s now too much hassle to read WP pages unless I do it on my laptop. Since I’ve probably just followed a link from google, that would mean emailing myself the link from my iPhone, and going to WP via my desktop. More wasted time. I’ll just stop using wikipedia, thanks.

So far this morning I haven’t been able to access WP short of manually changing the URL to go to a country-specific Wikipedia mirror, switching to a “slow” (non-broadband) internet connection, reloading the page, and hitting the stop button before they redirect me to a “cut down” version, and no link to escape from it. There’s a link for you to “comment” on the new “feature”; my commentary would have been unprintable, so I declined.

Gamespot: we don’t want money, money is for wimps

The other week I noticed that Gamespot – one of the big ad-driven news + reviews/cheats/etc websites for games – is still locking-out all mobile users. That’s probably a fairly substantial load of ad revenue they are literally throwing away every day.

The web, HTTP, and HTML…

Why do people do this? I don’t know. But here’s a few points you should bear in mind:

• No website should ever block content based on the user’s device
• No website should ever have a flash-only front page
• Since the very first versions of HTTP and HTML in the mid-1990′s, the web has been designed to avoid these problems; this shouldn’t be happening

Content Blocking

Gamespot checks your web browser when you fetch any article, review, etc. If it finds you’re coming from an iPhone, then it refuses to let you view the content. Instead, it serves up a custom “news page” that is identical no matter which link you came in on. There is no way for you to see the actual content you tried to view – literally: they do an auto-redirect that wipes it from the URL.

I can see no reason for this other than the bizarre assumption that an iPhone was launched 10 years ago with a tiny black-and-white screen and an inability to scroll and render web pages. I would love to ask the Gamespot web design team: have you ever seen an iPhone? You do realise it has a better web browser than most desktop PCs, yes? So … why are you manually blocking them from your website?

Amazon has for a long time done a similar thing with any mobile device (again, sadly, the stupid bit is that they apply it to devices where it’s completely unnecessary) – except that Amazon has three essential features which Gamespot lacks.

Firstly, they do actually show you some of the content you were trying to view (not all of it. ARGH!)

Secondly, there’s always a link on the page to view the real version of the page. If you click that, it gives you a warning something like: “YOUR MOBILE PHONE MAY NOT RENDER THIS PAGE … ARE YOU SURE!!!!????!”. Of course, this is somewhat inappropritate when applied to most smartphones, especially iPhones. But hey – at least the option is there.

Finally, they have a link something along the lines of: “Do you want to permanently stop seeing the broken, cut-down version of pages on amazon.com? You can re-enable them whenever you want”.

Irritating, patronising, and foolish (the default should be “view the website normally”, not “don’t view the website”) – but at least you only have to fix it once, and you never again get problems. Gamespot et al offer no such option – they just block you, dead.

Flash-only front pages

About 50% of indie studios have decided to put a massive flash on their front page, most of them with *no* link to “skip intro” or “go to website” or any kind of navbar. About 50% of them (in my sampling over the past few weeks) have made that flash NON clickable: you cannot (you are “not allowed to” ?) view the “real” website until the flash has loaded, you have seen the self-promoting advert for the studio embedded in it, and clicked some internal link at the end. This was foolish, unnecessarily slow, and contrary to the spirit and standards that drive the web even 10 years ago when it first started happening.

Games industry companies please take note:

The 1990′s phoned – they want their web-designers back.

(real web companies don’t do this kind of thing any more)

But now, with the iphone, it’s particularly dumb: it is de-facto content blocking – because the iPhone cannot / will not run Flash. If the Flash is clickable, you can at least (if you know what the studio did – which many people won’t guess) access the site anyway. I’m amazed how many sites don’t even give you that small fillip.

If this post persuades JUST ONE web designer, somewhere, to wake up and smell the roses, and spares us yet another self-blocked website, then I shall be happy.

Of course, maybe I should be grateful that we’re even this far “ahead” … I heard from someone the other day that he still has to explain to web design teams that websites don’t need to be hardcoded for rendering at 800×600 any more (i.e. that – OMGWTFBBQ! – everyone has rather larger desktop screen resolutions than that these days; or else so much smaller that hardcoding to 800×600 isn’t going to help at all).

Filed under : Web 0.1, bitching, games industry, iphone, marketing

9 Comments »

I just discovered something rather annoying: people have been leaving voicemail on my cell phone. Please don’t. I can almost guarantee that I’ll never receive the message.

Normally, this doesn’t happen: I have a recorded message explaining:

• don’t leave a message – I won’t receive it. No, really. Don’t leave one “just in case”: you’re being self-delusional. I WILL NOT receive it. (It’s amazing how many people struggle with this concept :)).
• if you need to contact me, you should already have my email address; if you don’t even have my email, then you shouldn’t have my phone number in the first place
• if it’s urgent, you can try sending an SMS, but if I wasn’t answering the phone, it’s probably switched off or out of cell reception, so I might not see the SMS for a while either (but at least I’ll see the SMS as soon as I’m back in range / switched on)

(more…)

Filed under : bitching, maintenance

3 Comments »

The title says it all really; for whatever reason, the Thunderbird developers appear not to have tested TB on OS X. So much of the basic functionality doesn’t work in the latest beta – this isn’t even alpha-quality code (on OS X). I’m sure it works fine on Windows (or else you’d have thousands of people complaining long and loudly).

I had this suspicion with Shredder 2 (the last alpha), where basic features – like sending emails, and viewing messages in a folder – would regularly crash the OS X build. Even for an alpha that should have been unacceptable, or fixed very rapidly. Where’s the regression testing?

But I hoped I was just being cynical, and so I moved on, and forgot about all that. My experiences over the past couple of months with the beta have recreated that suspicion, and cemented it. For instance, I lost a couple of hours of work today because TB on OS X has major bugs in its synch code. I watched as it silently deleted all the changes from the activity manager. No errors. Nothing. Just … gone. Even without ever having read TB source code, I can think of two or three obvious coding errors that would cause such behaviour, and none are things I’d exepect to get into a project as popular and well-known as TB.

So … what gives? What’s wrong with the OS X builds of TB? Why are they so very, very bad? Why do they have so many dataloss bugs?

Sigh. At least I can fairly rapidly re-do all the work I lost. Time to start looking for a new email application. Maybe I can find a version of Mozilla Mail that still runs on OS X? (FYI: Mozilla mail was the thing that Thunderbird was based on / supposed to replace. Unlike TB, it actually worked. It was faster, had more features, but looked a lot uglier. I’ll happily sacrifice “good looks” if it gets me “supports basic email features from 10 years ago”)

Filed under : Web 0.1, bitching

17 Comments »

I had to do some iPhone prototyping recently, and we had a trial copy of Unity to hand. I thought this was a great excuse to try using it. First impressions of the editor/IDE/environment – at least on OS X – are not good.

NB: In general, in terms of what can be done with it etc, I’m a fan of Unity. But I’ve never developed with it directly myself, and I’m now finding it surprisingly painful / steep learning curve.

Need to know basis

None of the built-in tutorials work, flat out, because the startup code has apparently changed substantially since they were written. The tutorials keep talking about things like “create a new project; by default it will X and Y and Z” but Unity no longer does any of those by default. Sadly, the tutorials don’t tell you how to get any of those manually – because, you know, they’re done for you by default, why would you ever need to know how to do them by hand?

File Association Theft

I was also *extremely* unhappy to discover a short while later that Unity has stolen the file association for PHP files. Under OS X (thanks, Apple) managing file associations is a surprisingly irritating business, as bad as with Microsoft Windows (Apple deems users too stupid to be allowed to simply edit associations – but applications are allowed to overwrite each other with absolute trust from Apple, and no user intervention allowed), so this is a pain to fix. In particular, I have an entire *suite* of applications and IDE’s for doing web editing, including a specialized high quality PHP IDE. Not any more; Unity has clobbered that with a crappy text editor that does nothing more than basic syntax hilighting. This is pretty offensive: firstly, don’t steal my files without asking, and secondly – give me back my IDE!

NB: I have no idea how it has done this, but Unity appears to have overridden OS X’s systems for file association management – following the standard procedure (e.g. here) has no effect, and Unity keeps stealing control of the files immediately that you confirm you want to give the assocation to some other app.

At this rate, if I can’t find out what it’s done to my OS and undo it, I’ll be uninstalling and deleting Unity with extreme prejudice in the very near future. Sure, this is partly Apple’s fault for assuming all apps are perfect and all users are not, but at a simpler level I just cannot afford to have a non-functioning development computer just because of one badly behaved application.

Filed under : bitching, dev-process, fixing your desktop, iphone, programming

10 Comments »

We finally have a mainstream response (as opposed to responses sitting quietly in obscurity in the IGDA.org forums) to the issue of IGDA Board Members pissing all over IGDA’s main tenets:

http://www.igda.org/newsroom/memo_032209.php
(from the IGDA monthly newsletter that just went out to all 15,000+ members)

Instances of the following words in that response:

sorry – 0
mistake – 0
fault – 0
acceptable – 0
unacceptable – 0

I guess that says it all, really. If that was meant to be an apology, or a recognition that the board and the org did *anything* wrong here at all, then it’s a clear FAIL.

If not … well, what was the point?

The fact that the org refuses even now to accept that it did anything wrong, refuses to apologize, or to commit to acting differently next time, that there is no action item offered to rectify it taking 4+ months for the organization to respond (having only acted *at all* after it was pushed into the news, DESPITE the board being present at the damn event), well…

All that seems to me to say:

“it’s business as usual, folks; EA, Epic, and all the other abusive studios – don’t worry, IGDA has your back! Our members come second”

I didn’t used to believe that – I know many key people in IGDA personally and knew they would never ever espouse that – but I’m now staring the reality in the face and seeing that “the organization as run by the board” != “the diligent and wonderful individuals I know who contribute so much”.

PS: quick shout to Tom Buscaglia (one of the board members who’s so far apparently done nothing about all this) – if, after reading this post, you’re going to accuse *me* of being “a whiney little bitch who would rather quit after a loss than jump in deeper” again, then I suggest this time that you link to the post URL rather than the blog domain. That way people reading this will get an automatic trackback link to your site, and get to see your response this time around :).

Filed under : bitching, games industry

3 Comments »

(seriously – otherwise I’ll be phoneless thoughout GDC :( )

I’ve had no interest in cracking my iPhone, so I haven’t.

Until I discovered the other day that my incompetent network (O2) won’t allow me to make calls in the USA on the agreement I have with them, so I need to use a local USA SIM while I’m there. Unsurprisingly, all O2′s own staff openly advised this as the only sane course of action. They were apologetic that this was necessary. O2 loses nothing if I unlock the phone.

And then I discover that Apple’s undocumented 2.2.1 update which I was bounced into installing has disabled all known unlock processes except for the hardware ones. If I had bothered to do the unlock a few months ago, it would have worked perfectly. Now, with less than a day until I leave the country, there’s nothing I can do.

Help? Anyone?

(NB: I’m not on a contract. I’m not even registered with the network. I’m sure the EU commission will sue Apple’s ass over this sooner or later and force them to stop retailing locked phones in the UK. That is of no help to me *today*)

(NB2: Apple’s lack of respect for their consumer continues to impress me every year. I know *why* they did it (network operators forced them to), but that doesn’t excuse screwing the consumer without warning them what you’re about to do to them. Undocumented updates are vicious)

Filed under : GDC 2009, Web 0.1, bitching, conferences, funny, iphone

5 Comments »

Today, I attempted to login to Feedburner.

I had to go through not 1, not 2, but three different adverts for “here’s why we’re forcing you to create a new Google/Gmail account, you really want to do it, it will be good for you”.

The third one has a link “ignore this and login to feedburner anyway”.

That link redirects back to the adverts. It didn’t last time I logged in, a mere week ago – back then, it did what it claimed to do: logged me in to Feedburner.

This time, I clicked a dozen times, with waits in between, just in case it was a momentary glitch.

Nope – Google/Feedburner is engaging in some kind of sadistic schadenfreude designed to make Feedburner users collapse psychologically, and give-in and create Google accounts. Or, if like me they already have a Google account, to delete their Feedburner identities and throw away privacy, personal data, and any semblance of control of their online personas to instead merge their account IDs.

(in some kind of surrealist nightmare, the final ad opens with the words “Moving to a Google Account is easy” (no, actually, it’s enforced), and finishes with the line “You don’t have to move today, but we’ll remind you each time you visit feedburner.com” (um, no – you’ve placed an auto-redirect that does, in fact, force me to “move today”. Bastards)).

So. Feedburner? Google? You say: “F*** YOU”? Well, then, I say: “F*** you too; harder”.

Filed under : Web 0.1, bitching, maintenance

9 Comments »

Disappointing

No spam filter (there is one, allegedly, but it’s invisible, non-configurable (!), and missing obvious spam words)

(there are comments on the official forums from 3 years ago saying “we need some basic anti-spam tools” and the author replying with “we’re working on it”)

Annoying

No spam controls for reactive processing (banning users, marking comments as spam, etc).

Unforgivable

Each comment can ONLY be deleted one, by one, by one, requiring 6 mouse clicks per deletion!

The concept of a list of comments, and a checkbox next to each, and a “delete selected comments” is now well over 10 years old – and it only takes a few minutes to implement with PHP + SQL. If you ever find yourself making a web app for general usage, please don’t forget this core feature :).

(I’ve been too busy for the past month making actual iPhone apps to finish my free PHP FAQ platform, but watch this space, shouldn’t be too long now…)

Filed under : Web 0.1, bitching, programming, web 2.0

5 Comments »

I can no longer develop iPhone Apps. I am on my eighth attempt to download the 1.75Gb 2.2.1 SDK – without which, XCode refuses to even talk to my iPhone any more, because I allowed the iPhone to upgrade.

EDIT: I have it! I HAVE IT! YES! NO MORE PRAYING TO THE GODS OF ****ING APPLE’S CRAPPY WEB SERVER! (it’s probably a corrupted download)

The problem?

Sheer mind-numbing incompetence by Apple’s web team, it would seem (NB: I may be wrong; I haven’t tried packet-sniffing the HTTP traffic to prove this beyond all doubt, but my experiments with different browsers and different ISP’s strongly indicate it is the case). They’ve (mis)configured the webserver over at developer.apple.com to kill partial downloads of this monstrous file.

If you cannot download the whole thing before Apple automatically logs you out of their website (which they do every half an hour or so if you don’t actively surf the site), then your download is cancelled, server-side. You cannot resume from where you left off (their web server refuses to honour the HTTP command that tells it to resume a partial download).

What did they do wrong?

It’s a session-management problem – you CAN pause and resume a minute or two later. Just not half an hour later.

The other major bugs in the Apple websites (including that force logout!) suggest that some novice web programmer made a half-assed Session-management system that, well, sucks. (am I sounding angry? Hell yes; Apple’s website is behaving like some naive bricks-and-mortar company from 1999, not 2009). Or maybe they picked up a 3rd party one … that sucked. Whatever. It seems the session-manager is overriding Apache’s built-in support for this core element of HTTP, and saying “no”. For no reason other than bad web coding.

In passing, I noticed that they are *still* (allegedly – it’s easy to fake) running the same version of the Apache webserver that they’ve been shipping by default with OS X Server for some years now – a version that is more than 4 years old, 8 versions behind the current “maintenance” release, and 1 major version behind the “mainstream” release. Personally, I wouldn’t run Apache 1.x if you paid me, not with 2.x out there and running a “proper” webserver arch (apache 1.x is not a real webserver, it’s a hack using forks that makes it unnecessarily slow under heavy load).

</rant>

Filed under : Web 0.1, bitching, iphone, web 2.0

0 Comment »