I read far too many feeds I’m sure. This morning I’ve spent literally about half an hour catching up with all the feeds I’m subscribed to and the last time I was reading feed items was last night at about 9:00pm. Google Reader only was able to tell me that there were ‘100+’ new items but I’m sure it was at least 200-300. In one night (for me, anyway).

Now when I see a feed item that isn’t in full in my Reader, I am forced to click through and open a new tab to read that story. When you consider how much time I have to spend to keep up anyway, this is very irritating and breaks the flow of reading.

FullFeeds is an online petition (JavaScript mandatory) with the intention of getting content publishers to use full-length feeds to avoid this irritation.

I signed it, saying:

The BBC are very guilty of this – and for no real reason. In the UK, they have no ads (and shouldn’t do) because they’re funded by the TV licence. So why do they have only extracts in their feeds?

If a feed isn’t in full, I’m less likely to read it in full and less likely to ‘star’ it (where starred items show up on my homepage on the right).

I don’t even mind seeing a small text (preferred) or an unintrusive, small graphical ad at the bottom. I understand the need for content publishers to monetise their content and I don’t mind that (provided of course it’s not a massive great banner ad, you get the idea). But only publishing part of the feed content and forcing me to click through (and then consume the ads) is a terrible idea.

In fact, I’m much more likely to accept advertising via RSS. I use NoScript for Firefox, which means that JavaScript is turned off for sites I don’t specifically allow. (I’m paranoid!) But seriously, I don’t trust random sites running JavaScript on my machine without clearing it with me first. This means that advertisers that rely on JavaScript to deliver their ads don’t get their ads to me. And tough luck, because in my opinion, they shouldn’t rely on JavaScript for that purpose.

Since ads in feeds don’t use JavaScript, I’m much more likely to see them and accept them. Hey, it’ll work before we start seeing ‘AdBlock’ extensions for feed readers…

Content publishers – take note and enjoy ‘feedvertising’ while it lasts.

… and why Apple can’t be bothered to properly localise Apple keyboards this side of the Atlantic.

But I should really start at the beginning. Basically, for a long time now I’ve been pondering a decision to buy a brand new MacBook. I came to the conclusion that I will, but I’ll wait for not only Mac OS X Leopard, but also hopefully for an Intel Core 2 Duo (Merom) version, fingers crossed.

So today I saw a MacBook ‘in the flesh’, not for the first time, but I suddenly noticed that the keyboard layout was rather strange. Basically, British layout keyboards have some minor differences to US ones, including having a £ sign located at Shift+3, and having the “ character at Shift+2, not Shift+’ (the @ is then located at Shift+’ instead). Plus, also the # sign is somewhere else… you get the idea. It’s standard QWERTY with some tweaks.

Now the MacBook (and every other Mac in the, ironically named, PC World store) had the £ key in the correct place, but all the other keys were US style.

This would be a major problem for me, as I’m so used to typing with the British layout, I’d get endlessly frustrated having to hunt around actually looking at the keyboard to find a key. In fact, I couldn’t see the hash key (#) anywhere on the Apple keyboards! Not that I wouldn’t get over switching to the US layout, but you see, every other keyboard over here is in the standard British layout described above. I do not want to learn two different keyboard layouts and have to think which one to use!

A look in System Preferences revealed that the MacBook believed it was using a British keyboard layout and I couldn’t find any other layout among the many localised keyboard setups that seemed to offer the proper setting. Imagine if I plugged in the keyboard I’m using now. Like 99.99% of keyboards over here, it’s in the British configuration. I’d press “, but end up with an @ – I’d end up getting very frustrated.

Thankfully, I’m not the first to discover this and Phil Gyford has devised a solution which involves installing a custom keyboard layout. Thank you, Phil! Without this I may well have had second thoughts about my MacBook purchase, which I’ll make as soon as Leopard is out (provided they’re not just about to release a Core 2 MacBook at that point).

It won’t fix the labelling of the keyboard, but to be honest I won’t be looking at that while typing, will I?

Small details. Big impact. Like having the wrong keyboard layout that almost made me think twice about getting a Mac.

Seriously, sort it out, Apple. If every other manufacturer can do it (seriously, every computer sold in the UK gets it right), then surely you can.

You almost scared me off your entire product range. And you were saved only by a hack by a user of your products.

OK, end of rant.

Recently Debian Linux project have come under fire from Mozilla (the parent organisation of Firefox and Thunderbird) about using the ‘Firefox’ name in Debian.

Basically, the Debian Free Software Guidelines state that Debian can’t use the Firefox logo (as it’s not under the same licence as the Firefox code and the licence isn’t considered Free). So for some time, Debian (and derivatives, such as Ubuntu) have been using Firefox with a custom icon and calling it ‘Firefox’, not ‘Mozilla Firefox’.

Apparently, this isn’t acceptable any more and Mozilla want to either see and approve all the changes Debian make to Firefox, or see that Debian and derivatives stop using the Firefox name.

I think this is ridiculous. How can this be helping free and open source software if we have squabbles amongst the different communities?! If Firefox is removed from Debian/Ubuntu/MEPIS/all the others, then it’s not exactly going to be easy to convince potential Windows switchers that Linux is better if they can’t use Firefox on it (that they might use on Windows already)!

Please, drop it, Mozilla and let them keep using the name. If that can’t happen, then Debian and Ubuntu should just rename it ‘Web Browser’ or something like that. In my opinion, having Firefox in Linux (remembering Ubuntu is the most popular distro) is very important.

Of course, if it does get removed, there’s nothing stopping people from downloading vanilla Firefox from the Mozilla site, but that would be a pain. Firefox, in my opinion, needs to be built in to the OS. Firefox is usually the experience the average user will have had with ‘open source’, so keeping that consistency is vital.

Please don’t do this, Mozilla (not that it would affect me personally, being a Fedora user)…

Couple of other links on this topic:

Chris Van Patten has got a simplified ‘conversation’ version of the story and Jacob Peddicord was involved in the discussion and has a more detailed look.

Quick update – if you remember, I had problems getting Windows 3.11 for Workgroups running on top of FreeDOS in VMware (wow, long sentence).

Well, a kind friend donated me an ancient copy of MS-DOS 6.22 (on four floppy disks, may I add) which has helped me actually get WfW running inside a VM (thanks James).

It involved a bit of hacking, but now I have WfW running inside VMware.

Basically, the MS-DOS install goes fine, but on rebooting I found that it wouldn’t boot – the PC speaker started making a horrible noise and garbage was printed to the screen. If you press F8 inside the VM while booting, you get the option to control what happens at boot up. Answer no to all the questions, and you will boot into a DOS environment (albeit limited to 640 KB RAM).

Then insert the WfW disk, and run:

> A:
> setup

Then run through WfW setup as normal. Now when you restart, you need to press F8 again while booting MS-DOS and answer the questions as follows:

DOS=HIGH [Y,N]? Y
DEVICE=C:\\DOS\\SETVER.EXE [Y,N]? N
DEVICE=C:\\WINDOWS\\HIMEM.SYS [Y,N]? Y
COUNTRY=044,,C:\\DOS\\COUNTRY.SYS [Y,N]? Y
DEVICE=C:\\DOS\\DISPLAY.SYS CON=(EGA,,1) [Y,N] Y
FILES=30 [Y,N]? Y
LASTDRIVE=Z [Y,N]? Y
DEVICE=C:\\WINDOWS\\SMARTDRV.EXE /DOUBLE_BUFFER [Y,N]? Y
DEVICE=C:\WINDOWS\\IFSHLP.SYS [Y,N]? Y
STACKS=9,256 [Y,N]? Y
Process AUTOEXEC.BAT [Y,N]? Y

After this point, answer yes to the rest of the questions (your country line may differ slightly or be absent completely depending on which country you chose during MS-DOS setup). Now you can run win from the command line to boot into Windows.

Now you can open Notepad (in the Accessories group) and edit C:\CONFIG.SYS. Remove the offending line
DEVICE=C:\DOS\SETVER.EXE. (In fact you could do this using MS-DOS’s edit, but this is presuming you want to install Windows anyway).
Now MS-DOS boots perfectly and I can be nostalgic with Windows 3.11! Remember, if you add the line win to the end of file C:\AUTOEXEC.BAT, Windows will start on boot.

WfW screenshots? Oh go on then, they’ll be here soon. UPDATE: See the whole set.

I’ve got the new hard drive now (a Western Digital) and I’ve decided to rebuild Windows XP on this machine (but not Linux – I’ll copy that partition over later).

So I spent quite a lot of last night installing XP, updating XP, getting security software up-to-date and wrestling with passwords until they worked.

I’m actually practising what I preach, and I’m no longer going to be running as administrator all the time on Windows (running as admin is one of the reasons why malware can be so damaging on Windows). Instead, I’ve got an admin account and my normal account and I’ll have to get used to ‘running as’ applications that are too stupid that they need admin privileges.

It is a real inconvienience in XP, but it’s generally a good practice to employ when using a Windows machine. Running as admin is like running as root on Linux all the time – I would never even think of being root all the time on Linux!
So, still quite a lot of software to go, I’m afraid, and things to configure, set up, wrestle with…

My first 160 GB hard drive (made by Maxtor, for the record) just messed up again and it’s essentially decided that there’s no boot sector on it.

Now this has happened many times before, so I’m not too worried for the data. Usually, you just switch the PC off for an hour or so and it works again. Needless to say, I’ll be backing everything up asap (if I can boot back in, that is). I also have last month’s backups from about the 13-18th – can’t remember exactly. Why is it that they never fail right after a backup??

So I’m bringing you this post from my server/development machine, which is, ironically, running second-hand hard drives thrown around from machine to machine until they made it here.

Oh, and we have a 75 MHz Pentium 1 lying around here that has hard drives that are probably close to or even over 10 years old. Those hard drives (1.2 GB and 800 MB HDs) are still running without a hitch. I suppose it’s the high capacities that bring the low reliabilities…

This Maxtor is a couple of years old, by the way – and a replacement for the first one I got, which was so faulty Windows wouldn’t install.

OK, rant over.

EDIT: About 30 minutes later, it boots fine, and I start backing up. Sigh.

As part of my feed reader quest, I’m trying out Google’s Reader, which is a bit more fully featured than Personalised Home. I’m going to use it for a few days and then post a review at Gizbuzz about my thoughts.

In the meantime, you can see which feed items I’ve found interesting below:

(Apologies to anyone who came to an old, broken version of this post via the feed – I messed up the publish somehow and had to pull the old post and rewrite it)

A lot of WordPress-powered blogs recently got a load of comment spam. Nasty. In fact, I had to do a bit of a clean up job (and so did many people using WordPress), all because the Akismet spam database appeared to be down.

Akismet is an awesome service for WordPress-powered blogs which checks your comments against the spam database and either puts the comment in the spam bin, the moderation queue or straight onto the post depending on how similar it is to the spam.

Unfortunately, when the spam database goes down, as happened recently, all comments get let through … with disastrous results. So I’d like to take this opportunity to thank the Akismet guys for their great service; the fact that I don’t notice the spam when it’s working means Akismet is great.

According to their site:

One of the reasons we’re doing Akismet is we’ve built up a highly fault-tolerant infrastructure that can handle huge amounts of traffic and processing. However if something ever does go wrong your comments will simply go into the moderation queue.

Not this time, apparently.

But we love Akismet! (especially when it’s online!). No, seriously, we do. We love WordPress too.

UPDATE: The Akismet blog has an update on this – it seems to be back online and working for me. Apparently some new code broke the API, but it’s working again now. Thanks guys for the great service. Thanks also to Antony Pranata for (indirectly) leading me to this update (and linking to me in the first place!).

This tutorial which shows you how to build a KDE application in C++ using Kdevelop and QT Designer is very good. As part of my attempt (can I emphasise attempt here) to learn not only C++, but to be good enough to build a KDE application, I tried it.

It’s very well written, but I think I strayed a little too much off the course of the tutorial and I ended up being unable to build my application. Something about some error somewhere. Still, there is source code for a version which does work, so I might take a look at that and see where I went wrong. I swear I did something wrong with the Automake Manager… Ah well, I’ll have another crack at it later and hopefully I’ll be able to build a KDE application Sometime Real Soon.

Hey, at least I’m making progress, albeit slow progress. This post comes to you courtesy of my SUSE 10.1 software development installation.

Windows for Workgroups 3.11 was the first version of Windows that I used, and I thought I might have another look at it, just for the fun and nostalgia value.

But I wasn’t going to install it on a real machine – I have setups I don’t want to lose! So VMware was launched and I had a try getting it to work on a virtual machine.
Just a word – the original machine which ran WfW 3.11 no longer has it installed; not that Microsoft would be too worried about multiple machine uses of an essentially dead product. The machine originally came with IBM PC DOS and OS/2 (neither of which now live on it either).

Initially, I was going to put PC DOS on the VM first, because Windows at this stage wasn’t really an OS in its own right, it needed DOS to run the setup. However the IBM PC DOS floppies weren’t looking too good, and the PC DOS install had a hiccup. It completed, but important stuff like himem.sys (deals with extended memory, so you can have over 640k RAM) weren’t installed, so it wasn’t much use.

So instead I downloaded FreeDOS, the free and open source DOS system, that claims pretty damn good compatibility with MS-DOS. I had to find a mirror that worked, and a version that worked, but when I did, I successfully installed FreeDOS on my 200 MB virtual hard drive.

Now, to the Windows install – with screenshots (click any one to make it bigger, thanks Zooomr)!
» Read the rest of this post…