Category Archives: General

Google Hates Me

Note: This entry has been restored from old archives.

Well, at the very least Google has decided that I am not dissimilar to a small and rather stupid perl script:

Google hates me

What makes it really annoying is there there seems to be some sort of unhateme option where you type in the random characters from an obscured image but the image of the characters times out! I can’t be unhated, whaaa! It only happens for my personalised Google home, the normal Google search works fine.

No contact info provided, just a sad little note apologising to me if I happen to care (i.e. if I really am not a script) and hoping to see me again.

It works okay if I bounce through a proxy in another location. Maybe the site I’m at has a Google bashing infestation and has been blacklisted… now that would be amusing.

Web Album

Note: This entry has been restored from old archives.

I’m giving serious thought to moving over to an online service to keep photos. For several reasons I don’t like the idea though. They’re my photos, I’m a bit iffy about trusting them to some corporation to store and display. Control of the look-and-feel of the sites that provide this service is limited. The look-and-feel of the existing sites is just plain awful.

What might change my mind is Picasa Web Albums. I foolishly trust Google with a lot of my data so why not some photos? And while it is unlikely that the look-and-feel will be highly customisable there is a good chance that the default will not suck.

My main problem with my existing photo publishing approach is that the barrier between taking photos and getting them to the web is too great (thus I never get around to uploading). I suck them off the camera, use one piece of software to browse-rotate-cull them and then another for touch-ups, then I upload them to Gallery and go through adding the captions (slow via web interface). Google has have Picasa for Linux (in Labs) and I assume that interaction between Picasa and Picasa Web Albums will be seamless, in typical Google style. When it all works I’ll give it a spin, and maybe my next album upload will be to Google.

The drawbacks are: It isn’t ready yet (beta for invited Picasa users), the Linux Picasa doesn’t support it yet (no idea when), there is only 250 MB of storage space (or am I willing to pay US$25 per year for 6GB of storage, depends on how good it is).

I wonder when we’ll have the Google range of home appliances? With a nifty web interface that lets you add ‘make coffee’ events to your calendar and have your Google Coffee machine obediently spit out a brew, and Google being perfectionists I just bet it’ll know what a Double Ristretto is. Like the white stuff in the coffee? Don’t worry, Google Coffee will interface with Google Fridge to ensure that it is kept in stock from your favourite online supermarket.

I await Google Baby, being able to upload your child to a Google server might make the concept of breeding more palatable. I’m sure a lot of parents wouldn’t mind their brats being reduced to little more than a feed on their Google home page.

Google Calendar

Note: This entry has been restored from old archives.

I’ve just discovered Google Calendar… maybe this will be the answer to all my calendaring dreams. Globally accessible, interoperable and seems to allow things like sharing events and inviting third parties to events. It is very new, but my “first 5 minutes” using it have been promising.

It would be good if Google put together a partner desktop app to go with it for local mirroring/synchronising of calendar data. Maybe one for the Mozilla calendar app?

Asynchronosity

Note: This entry has been restored from old archives.

I find this story about an asynchronous ARM processor very interesting. It immediately reminded me of two things.

The first thing was one of my Elec Eng lectures back in 3rd year for a course taken by one of my favourite professors. Asynchronous circuits were covered as something of academic interest, a method of building digital circuits that has many interesting properties but is ultimately just a curiosity. The problem being that there are too many complexities to make it a practical way to design ICs, designing a large asynchronous circuit is simply too difficult for any team of engineers, even very good engineers. These are the facts as I was taught them, as recently as 2002.

No doubt the content of the course lagged a little behind reality, since there must have already been people working on exactly this task and already building significantly functional asynchronous ICs. And now they’ve implemented an entire asynchronous ARM9 core!

I find this news remarkably exciting. It shows how rapidly technology continues to change and advance, that we’re continually breaking barriers that were so recently considered to be pretty solid. All thanks to the advance of technology driving its self.

And that brings me to the second thing. The theory that there is a phenomenal turning point in history ahead of us. A point that we can’t predict and beyond which we cannot imagine what will happen, I’d say we can’t be certain it will ever happen – we might not last long enough. It may even happen within the span of our very own lifetimes, that would be something worth seeing.

This is the point where our technology can autonomously design the next generation of technology, triggering a world changing chain reaction that the homo sapiens sapiens species as we know it cannot survive. Either in some way or another it’s the end of the line, or (more optimistically) we’re caught up in the process of driven evolution and become something we cannot imagine.

It really is a very exciting concept. It is just another crazy futurist theory of course, one I happen to find particularly attractive.

Big Iron

Note: This entry has been restored from old archives.

Well, I really mean bigER iron – bigger than a VPS at any rate. I’ve had enough of this whole VPS thing, performance is too unreliable (who knows how many VPSs they have on a system?).

So I’m likely to go for getting a dedicated server, since the thing is my primary communications hub (and also for a couple of other people) I think it is time to upgrade for the sake of reliability. I haven’t decided where yet, there’s good old EV1 of course who are fairly solid (though they do have an iffy reputation for getting IP ranges spam blacklisted and they don’t do Debian) but there are many other choices. It’ll either be in the US or UK since prices in these locations are still much better than back home.

It costs a lot more than a VPS of course, so ideally I’d like to share the cost around and form one of these “server collectives”. I have no idea if anyone is likely to be interested in this, but if you are send me an email and if I get any responses we’ll work out what we need and how much it is likely to cost.

The deal would be simple: If the server costs $x then each of ‘n’ users pays $x/n. For your $x/n you get a single shell, root access if you feel you need it (I think I trust almost everyone I know to admin a Linux system without major mishap!) and an equal say in the running of the machine. I’m happy to look after all server administration, my typical setup provides POP3 and IMAP with Courier, Postfix for SMTP, Apache for HTTP, webmail with SquirrelMail and I use ClamAV and SpamAssassin. Debian is the OS, I wont budge on that. And some packages are sourced from backports.org for the sake of keeping up to date (for items like ClamAV and SpamAssassin where being up to date is important).

As for the machine, that would be decided by the people involved. As an example, the cheapest EV1 option is US$99 per month (however EV1 only offer RHEL, so they’re actually an unlikely choice) and this is a 1.3GHz Celeron with 512MB or RAM and 60GB HDD. I’d think that having an IP per user would be sensible plus one for the system (for HTTPS use, system IP is for SquirrelMail), IPs are usually cheap (7 for “free” with EV1 for example). I think the basic HW specs with the EV1 example are fine for a fair number of normal users (assuming you don’t have an insanely popular website).

Now we’d (Kat and I would be 2 users of course) be willing to pay a reasonable amount (if nobody is interested then we’re likely to end up getting a dedicated server anyway), so having a small group of people with up to US$40 a month to spend would get us a decent setup – and the more people the cheaper the price is! Using the EV1 example then two extra people would get us the server for US$25 each.

Some dedicated server setups:

Me buying a server from eBay or something and just paying co-lo is also an option (some good options in the UK for around 50 GBP).

It may also be an option to have a cheaper class of user, for just email, FTP and webspace say – but it’d be easier if everyone was equal. An equal share user can have as many web domains and email addresses as they like of course and can provide simple hosting for others (for example I host a domain for my Mum and handle her website and email) but the only login access she has is for email. So long as the setup of a user does not adversely affect others on the system I’m sure everyone would be happy, so really the only restrictions are that only paying users get a shell and root access.

So send me an email if you’re interested!

Strange But Universal(?) Editor Feature

Note: This entry has been restored from old archives.

I was wondering why some of my results were going wrong when doing some pattern matching in Perl. I have a file that gets generated by a script and has one big RE on a line that is read into another script and executed. For some reason the RE doesn’t match what it should… just then I accidentally discovered that it does match and then doesn’t match if I have had the file open in vim, so even though all I am doing is opening the file and then immediately doing a :wq (should just be a :q then I wouldn’t have had this problem, but I do also need to tweak the RE a bit at times) the file content is modified!

What is weird is I then tried the same thing in nvi and then in pico and they do the same thing! At this point I’m checking the difference using md5sum. Each editor spits out the same slightly modified file?!?!

What’s the difference? The file ends up one byte longer… all three editors add a ‘n’ to the end of the file without being asked to.

At first I thought it was something to do with line length (my RE is over 400,000 chars long) but the editors do this for files with a single line only three chars long too… so why do they do this? If three out of three editors tested do the same thing it indicates that it might not be a random decision. As far as I know the editors have completely different codebases, so it isn’t just a shared bit of buggy (unless something in a shared library is doing it). I have a suspicion that there is a reason for this and I wonder what it is!

For now I’ll just add the chomp to my Perl script (it should have been there in the first place) and seek enlightenment later.