software · Grey Nicholson

Self-driving cars: software crashes

2018-02-25T23:28:00+00:00

No car is self-driving.

A “self-driving” car is piloted by software, which is ultimately written by a person. You don't know who that person was; only that they were employed by a particular company.

They were probably sitting in an office somewhere in California when they wrote the code driving your car. Maybe it was 17:30 on a Friday and, despite caring sincerely about the work they were doing, they happened to be distracted by the prospect of going home. Maybe not. You don't know.

Do you trust that person with your life?

Well, the company hired them, so they can't be completely useless. You trust the company's recruitment procedures. …What are the company's recruitment procedures?

Anyway, presumably there are processes in place to review the code, and stop mistakes from making it into the final software. Presumably. You trust that there are, and that they work, and never fail.

Now imagine the company has made it illegal for you to see how the software works. Are you sure you trust this company with your life?

There should be a law saying that if a vehicle can be piloted by software, and it's capable of containing or hurting a human, then all installed software must be open source, and you must be able to prove that the source code corresponds to the software running in the car.

It has to be legally possible for the vehicle's owner (or prospective owner) to discover how their car might behave in a life-or-death situation, so they can decide whether they want to be responsible for the car's actions.

Responsibility

Logically, the manufacturer who wrote the software would be responsible, but they have no incentive to take responsibility for their cars' imperfections. Doesn't make money. Why admit your own flaws while your competitors keep schtum, look better, and rake it in? Any goodwill from better transparency will evaporate as soon as someone dies in an accident.

It's much safer to claim that the human pilot should have taken control at the critical moment. Capitalist governments won't argue with rich, profitably-taxable businesses.

Car makers will only be transparent about how their cars behave if they're obliged to by law.

Open source

Merely having access to the software's source code isn't enough. It must be legal to reuse the source code, for several reasons.

Morally, if Non-Specific Engines Ltd writes an algorithm that's better at saving lives than any other algorithm, shouldn't Acme Motors be obliged to used the safer algorithm in their cars, rather than forbidden?

Practically, you need software experts to audit the code. You want the code checked by an independent expert in the field of vehicle automation — not a business partner of the manufacturer — and that person will be a software developer.

If they use a similar concept in their own work later, Mom's Friendly Car Company could threaten to sue them, claiming they copied the code illegally. Software developers are rarely as rich as car companies; even the threat of a lawsuit would mean that in practice the code would go unchecked.

And again, morally, you can save lives here, by letting the developer reuse the good code.

Reproducible builds

Lastly, it needs to be possible to prove that the audited code is actually the code running in the car. You want an independent auditor to build the software for themself, in a development environment they trust, and get the exact same output as what's in the car. It must be possible to build the software reproducibly.

Otherwise checking the code is pointless — you still have to trust the car manufacturer, and you can't be sure the software's behaviour doesn't deviate in subtle ways in very specific situations. Maybe you don't care about any subtle differences, but maybe you do. The driver should at least be honest with you, and you can decide for yourself.

None of this will make sure a self-driving car is perfectly safe. All software has bugs. But at least you'll know the driver was acting in good faith.

Trade secrets and competitive advantage are not worth dying for.

…Or you could just trust the big friendly company… right?

Big New Tab Button

2009-05-06T18:09:00+00:00

In Firefox 3.5 there's a button on the tab bar that opens a new tab. It's a bit small. This add-on makes it bigger. That is all.

Remove Close Buttons from Tabs

2009-03-03T21:04:00+00:00

Having a close button on each tab in Firefox makes closing tabs discoverable, but at the expense of accidental clicks.

When the tabs get too small the close buttons on inactive tabs are removed, again as a trade-off between discoverability and accidental clicking, but there's still the off-chance.

So for those of us who're comfortable with closing tabs by middle-clicking them, Remove Close Buttons from Tabs removes the close button from every tab, all the time.

Hence the name.

You can get it from Mozilla Add-ons or, if you really trust me that much, directly from me.

It's licensed under the ISC licence:

Copyright © 2009, Grey Nicholson

Permission to use, copy, modify, and/or distribute this software for any purpose with or without fee is hereby granted, provided that the above copyright notice and this permission notice appear in all copies.

The software is provided "as is" and the author disclaims all warranties with regard to this software including all implied warranties of merchantability and fitness. In no event shall the author be liable for any special, direct, indirect, or consequential damages or any damages whatsoever resulting from loss of use, data or profits, whether in an action of contract, negligence or other tortious action, arising out of or in connection with the use or performance of this software.

Mmm… shiny simplicity.

Red Head points out that you could just set browser.tabs.closeButtons to 2 instead, if you'd rather use about:config than install an add-on.

Just Give Me A Back Button

2009-02-16T08:57:00+00:00

A while ago, the Mozilla developers did some research and related some anecdotes, both of which suggested that web browser users use the Back button rather a lot. So in Firefox 3, they made it bigger than the Forward button. (—except on Linux, where they imitated a native design instead, shrewdly surmising that (A:) we like our applications consistent, and that (B:) our native design is pretty sensible.)

At the same time, they combined the Back drop-down menu (showing pages you've previously visited in this tab in this session) with the Forward drop-down menu (showing pages you've visited in this tab in this session that you've then come away from by going Back).

A couple of years previously, I'd made two Firefox extensions that played around with these menus' appearance & functionality. One removed the drop-down buttons, aiming for a simpler look (and leaving the menus accessible by a right-click); the other turned each entire button into the menu trigger (also providing the simpler look, but working less efficiently without providing any extra capability). The latter was never supported, and when the menus were unified I stopped using (& thus supporting) the former.

A couple of hours ago I was thinking about how Google Chrome manages to reduce all of a web browser's UI into five toolbar buttons & two menus, and thinking about how to do a similar thing with other applications, particularly Firefox. I realised that I tend not to remember which page is one Forward, so when I want to go to any page other than one Back, I usually use the drop-down menu. So I may as well get rid of the Forward button.

I also realised I could. So I did.

Just Give Me A Back Button is named after Asa's memorably-titled blog post and consists of one line of code:

#forward-button{display:none}

—plus the necessary incantation to install it and apply it to the main Firefox window. It may look like arse on Windows & Mac, where the Forward button is visually integrated with the Back & drop-down buttons—I don't know. Don't complain. (Fixes are welcome, though.) It actually doesn't look like arse—I've now checked.

It mainly lives at Mozilla Add-ons, but if you trust me—and there's no reason why you should—you can also get it directly from me.

It's licensed under the ISC licence, which reads:

Copyright © 2009, Grey Nicholson

Permission to use, copy, modify, and/or distribute this software for any purpose with or without fee is hereby granted, provided that the above copyright notice and this permission notice appear in all copies.

The software is provided "as is" and the author disclaims all warranties with regard to this software including all implied warranties of merchantability and fitness. In no event shall the author be liable for any special, direct, indirect, or consequential damages or any damages whatsoever resulting from loss of use, data or profits, whether in an action of contract, negligence or other tortious action, arising out of or in connection with the use or performance of this software.

It's only a minor simplification. Arguably, those overwhelming menus are a much bigger problem (in every sense). But it's something.

How to set up a Canon i250 printer in Ubuntu 8.10

2008-11-26T18:48:00+00:00

Setting up a Canon i250 printer in Ubuntu is possible, but not automatic. Here's how to do it in less than five minutes.

Actually, it seems to be impossible. I've no idea what I've done differently, but the printer fails to work now. Cheers, Canon.

First of all, the reason any of this is necessary is because Canon is rubbish. They don't support drivers (software needed for computers to talk to peripherals such as printers) for any operating systems other than Windows and Mac OS X. They don't allow me to just give you a copy of their unsupported Linux drivers. And they refuse to release specifications that would allow other programmers to make good-quality drivers for everyone.

I won't be buying another Canon printer in the foreseeable future.

Anyhow, I've packaged up a script to install the Canon i250 drivers in Ubuntu 8.10.

This script will ask for your password so that it can install the necessary software; unless you trust me enough to give me full control of your computer, you shouldn't run this script without reading it first.

If you do want to run the script, you'll have to make it executable. To do this in Ubuntu, right-click the script file (install-i250-drivers.sh), choose Properties, and on the Permissions tab, tick Allow executing file as program.

You can then double-click the script file to run the script. It'll take a couple of minutes to download, convert and install the drivers so I suggest running it in a terminal.

If you prefer not to download a script file, you can paste the following equivalent commands into a terminal (again, don't trust me):

gksudo "aptitude -y install libpng3 libtiff4 cupsys alien"
cd /tmp
wget http://download.canon.com.au/bj/i250linux/bjfilteri250-2.3-0.i386.rpm
wget http://download.canon.com.au/bj/i250linux/bjfiltercups-2.3-0.i386.rpm
gksudo "alien --scripts bjfilteri250-2.3-0.i386.rpm"
gksudo "alien --scripts bjfiltercups-2.3-0.i386.rpm"
gksudo "dpkg -i bjfiltercups_2.3-1_i386.deb bjfilteri250_2.3-1_i386.deb"
gksudo "ln -s /usr/lib/libtiff.so.4 /usr/lib/libtiff.so.3"
gksudo "ln -s /usr/lib/libpng.so.3 /usr/lib/libpng.so.2"
gksudo "/etc/init.d/cups restart"

The script & code don't actually set the printer up. You will still have to do that in the normal way through System → Administration → Printing. When offered the option, tell the computer to use the PPD file at /usr/share/cups/model/canoni250.ppd.

It may or may not be necessary for the printer to be switched on while Ubuntu starts up, so if the test page fails to print, try restarting the computer with the printer on.

I created this script by mercilessly copying from Daniel Howard's blog post on installing the i250 drivers in Ubuntu 7.10—spot the difference. (Seriously, there is a crucial change.)

How To Make Scroll Bars A Little Bit Easier To Use

2007-08-05T23:17:00+00:00

When you have a mouse whose scroll wheel doesn't work, you tend to use the scroll bars a lot more. I usually click the little arrows at each end repeatedly as I read the page. This means that I'm not looking where I'm clicking, and so my mouse tends to wander slowly, which results in me clicking stuff I didn't mean to.

A simple way to fix this—and it applies to many elements of a computer's graphical interface—would be to move the mouse pointer to the middle of the button when you click it.

This should be done when the mouse button is lifted, and only to controls that afford repeated clicking. So menus and drop-down list boxes would be excluded, because just clicking them does nothing and they can be dragged to select an item (in Gnome, anyway). Buttons that dismiss dialogue boxes, and the Minimise, Maximise and Close buttons in windows' title bars would be excluded as well, because the buttons move or disappear after you've clicked them once. Tabs; buttons that bring up a new window that you're expected to interact with straight away; and radio buttons (“choose one of several”-type options) wouldn't be included, because you don't need to select them twice in a row.

But tick boxes would, as you might want to toggle them. Most toolbar buttons; program launchers (on Gnome's panels); those little up and down arrows next to inputs that want a number; toggles for expanding to show more information; and scroll bar buttons would be included, too.

Google Desktop 2 Really Annoys Me

2005-12-28T21:18:00+00:00

After a while I gave in and am now using Google Desktop 2, rather than Microsoft Desktop Searchbox Toolbar, to quickly find stuff. I don't, however, use the sidebar, mainly because it was battering the crapputer's CPU, and because its search box makes all of the toolbars (or “desktop flaps”) I used to have obsolete. And it's more zen.

This is why I don't use MSN Desktop Google (hang on...) - with GD2 I just type something and press Enter; if I want to Google my computer with MSN (you know it makes sense) I have to type something, press down, then Enter. Otherwise a results window is launched, showing the least relevant results first (or so it seems).

So Microsoft Google (take that, trademark!) is out of the question. Most of my pet peeves about GD2 are resolved by not using the sidebar, but not all. It turns out GD2 does have a hotkey - Windows+G or Ctrl+Alt+G; neither of these can be changed to my preferred F9, but I can easily AutoHotkey that problem away.

But there are still little, niggly annoyances: I can't have the search box on the left of the taskbar because it's re-added to the taskbar every time I log on - on the right. There's no need for a Maximise button next to the search field - the drop-down menu includes the option to display the sidebar.

It's fortunate GD2 doesn't display times very often, because Google insist on using twelve-hour clock-style notation, that hasn't been used by anyone real since the twentieth century (yeah, that's a long time ago); and there's no option to change the format. If you're going to pick one time format for everyone, rather than letting them pick (and this is actually a good idea because it avoids unneccesarily cluttering the user interface), and you refuse to use the format the operating system tells you to use, at least pick a non-stupid format and not the one your country (alone) uses.

GD2 doesn't seem to want to find the music I have stored in Shared Documents. I've specifically told it to index the folder, but it responds with “What? What are you on about? That doesn't exist. Seriously. Nowt there. Now go away.”. However it does return music files on my Desktop; or rather they were on my desktop for a few minutes and haven't been there for the past month. Guess where they have been... “There's nothing there. Really. La-la-la-not-listening-la-la-la-la!”

And it's really, obnoxiously American - besides imitating a twelve-hour clock. A normal “I'm Feeling Jammy” search for “news”, from google.co.uk returns BBC News; an equivalent search from the search box returns CNN. This despite downloading GD2 from desktop.google.co.uk - I obviously want searches done via google.co.uk, and not google.com. And there's no option to change this.

It's really annoying when 90% of something is damned good and the other 10% is imbecilic.

Google Desktop 2

2005-08-23T23:01:00+00:00

Well, that was good timing. One day later (even less actually - it was a measly 14 hours before the relevant remaindering), Google goes and releases Google Desktop 2. I have installed it and, very generally, I like it.

I like that it integrates with Firefox and Thunderbird by default and automatically. I didn't have to tell it my Gmail username and password for it to display incoming mail from Thunderbird. Apparently, the Web Clips panel adds subscriptions for websites I visit frequently in Firefox, automatically. (OMG they're trying to rename RSS WTF STFU!)

At the moment I use a custom native Windows XP toolbar. At the top I have a collection of frequently used files and folders, below which is the MSN Windows tooltop deskbar search text field jobby. Below that I have the contents of the My Recent Documents folder, i.e. twenty of the most recently used files and their containing folders.

GD2 can replace much of this functionality with its Quick View panel, and I can merge the rest with the toolbar on the left of my desktop. But, there are a few things that'll stop me from getting rid of my current setup and using GD2 full-time:

The sidebar can be auto-hidden, but it stays “always on top”; so the far top-right corner of the screen is no longer the currently maximised window's Close button, it's now just dead space on the sidebar.
When I point to the right edge of the screen the sidebar waits for a short while before appearing; this is obviously to prevent it popping up in front of a maximised window when I'm just trying to scroll, but that wouldn't be a problem if the sidebar didn't insist on being “always on top”.
When I click away from MSN Deskbar Windows Tool Search, it automatically dismisses itself and leaves me alone; GD2 persists, in front of everything else of course.
I've already got used to pressing F10 and having Windows Desktop MSN Bar focus itself; GD2 needs this option (and I have to be able to make it F10, not Ctrl+Shift+G or something similarly intricate).
I'd like to be able to see more than ten search results without opening a webpage. It doesn't matter how accurate the results are, it's not going to find all of a 13-track album. Desktop MSN Whatever still only shows 12, but the full-blown results window isn't a webpage.

There are a few other things I'd like to see fixed but that wouldn't necessarily stop me from using GD2:

Much of the news content seems distinctly North American; maybe it's actually just international. Maybe me looking at more UK- and Europe-related news would fix this automatically for me. But it should be a bit smarter and offer more tailored content (and fix the spelling of the UI) based on my location.
Panels should scroll automatically (where it makes sense), rather than just showing the first item
It still seems a little eager to fire up a webpage; there should be an option to turn off all webpage-based UI and handle preferences through proper chrome (or it should just do that anyway)
The search box is stuck to the bottom; maybe I want it at the top, or somewhere in the middle.
When I resize the preview box for news items (as I'm invited to by the grippy), I expect more content to be shown, not more whitespace.
When I type Maximo Park I also want it to find Maxïmo Park, like a Google web search would.
It'd be nice if there was a To Do panel that could integrate with my Sunbird iCalendar files, like Konfabulator's PIM Overview widget can. It'd be even better if it was an editor rather than just a display (only if it was properly interoperable with Sunbird, of course).
I'm not allowed to move the deskbar out of the taskbar and into another native Windows XP toolbar, like I could with GDS and can with Desktop MSN Windows Search Toolbar.
It'd be nice if the panels could automatically resize themselves to suit the amount of content they contain (as long as it happened while I wasn't looking).

So for now I'm sticking with Search Microsoft Desktop Windows Toolbar...box. If Google had been a day earlier, they might've just won.

Microsoft Desktool MSN Windows Search…top…bar

2005-08-21T21:55:00+00:00

Earlier today I installed Desktop Windows MSN Search (or whatever it's called) and I'm actually pleasantly surprised. I'd had Google Desktop Search installed for a while, but I've never found it useful.

Its results always seemed irrelevant. Of course, calling a desktop search program “Google” is something of a misnomer - Google is founded on the principle that “a link is a vote” and there are no links to inspect for local files.

GDS had two options to display the results it found: relevance and date. The former seemed very arbitrary, turning up files I haven't used in ages (such as archived files) over ones I use all the time (such as mqwtm.xml, The Big, Bad Source File).

The other option, sorting by date seemed backwards - it showed the most recently used matching file first - the one that I've just used and haven't had chance to lose yet; I'd only actually bother firing up GDS to find a totally obscure file.

Another thing - the firing up. GDS seems slow compared to Toolbar Windows MSN Search Desktop, because the former is HTML-bound. I don't want to have to load a webpage to find a file on my computer, no matter how fast Firefox is. And since the results are just links, I can't right-click and use the system context menu.

Windows MSN Desktop Microsoft, on the other hand, pops up a window showing incremental results as I type. Hang about - GDS could also do that. Oh, yeah - I turned it off because I don't like allowing Internet Explorer to render webpages, and Google's results are in the form of a webpage.

MSN Deskbar Search Tool does come with a lot of links to msn.co.uk inbuilt, but all of them can be turned off, and any search service (i.e. Google) can be used for the web search. With a few minutes' Options-massage, it can be adequately non-annoying.

Desktop MSN Windows' full-blown results window's interface resembles the Details view of Explorer, although it's clearly just an imitation - there's no context menu for the column headers. But it allows sorting by any of many criteria in both directions, including “relevance”. The preview pane is pretty useful in most cases, except when previewing a Microsoft Word format or even Rich Text Format document. Then it's headed with “You Need To Buy Microsoft Office To Look At This Properly Yes You Do Don't Argue No Wordpad Won't Do £££”. Seriously. Direct quote.

It's quite amusing that the icon for the inferior - well, useless - inbuilt Windows Search function thing appears in the results window's toolbar and can't be removed. Right Hand, meet Left Hand.

To summarise, MSN Desktop Windows Search Toolbar actually works. But Microsoft (or MSN or whoever's supposed to be responsible for it) do need to figure out what it's supposed to be and then assign it a name - one name, that it calls itself throughout. Like, everywhere. (Have I introduced you to Left Hand yet?)

Netscape 4 is newer than Internet Explorer 6

2005-04-12T05:03:00+00:00

Browsing W3Schools, I've found something interesting – IE6 is older than Netscape 4.

Sort of. From their incomplete and somewhat dogmatastic overview of web browsers: Netscape Communicator 4.8 was released from Netscape in August 2002. whereas [Internet Explorer 6.0] was released in August 2001.

I know this has nothing to do with the age of the rendering engine, but dumb users know and care nothing about the rendering engine; in their eyes Netscape 4.8 is newer than IE6.

So if you're gonna stop supporting one of them next time you design a website, shouldn't it be the oldest? (Ideally, screw both of them.)

It's a Weblog Entry!

2005-01-08T02:48:00+00:00

OK, so the normal service has been a bit thin on the ground. Aaanyway... I'm back off to university tomorrow (Sunday); any new text and/or other whatnot will appear at Mooquackwooftweetmeow B, my university webspace.

Meanwhile

It took them four years, but this Christmas the people in charge finally cottoned on to the idea of putting two and two together, where the first “two” is a lot of homeless people in London and the second “two” is an empty Millennium Dome.

Over in Mozillaland...

Some guys decided to call Firefox “1.0” for a change. It seems to have worked. Then some other guys did the same with Thunderbird; that also worked reasonably well. And then roughly 20 million people downloaded them and they saw that they were good. And they divided the Firefox and the Thunderbird from the other applications; the Firefox and the Thunderbird they called “cool!” and the other applications they called “less so”. And lo Internet Explorer became without form and void, and darkness was upon the face of its developers. And Bill said “let there be users” but there were no users, for they saw that it was bad. And the grace of web standards be with us all. Amen.

Or something like that.

And in Operaworld...

They made a browser that can talk like an American, but it still insists on trying to sell me things I don't want, and I can't stop the browser or webpages from doing it. I guess they're firmly targetting users who can't see.

Oh! And...

It's 2005, you know - happy new year to everyone.

Jukefox

2004-09-13T16:43:00+00:00

It'd be nice if more extension authors read the documentation all extension authors must read and learned about the Firefox Version Format. A lot of extension authors seem to think 1.5 > 1.45. This would be true if Firefox version numbers were decimal numbers... but they're not - they're a string of integers, each separated by “.”. So 1.5 is indeed “one point five”, but 1.45 is “one point forty-five”, forty-five is greater than five (no, really), so 1.45 > 1.5. Of course, 1.5 > 1.4.5...

I suppose it was smart to market Firefox 0.10 (whose release is imminent) as “Firefox 1.0 Preview Release”, even though some folk seem to think it's not a milestone... which it is.

Meanwhile, with crazy autohiding controls and drag-'n'-drop placement - FoxyTunes 0.61 turns out to be a highly polished extension.

Still, they should've called it “Jukefox”.

IEBlank

2004-09-13T14:39:00+00:00

The Internet Explorer crew over at Microdollar clearly spend a lot of time thinking about browsing the web - as well they should, since they're supposed to be making a web browser. Now, they've pooled their collective thoughts into the IE weblog's Browsing the Web section, which serves as a useful, entertaining and enlightening glimpse into... oh, hang on a sec... what's...? No, it's definitely empty.

Odd.

Firefox extension wackiness

2004-08-24T22:29:00+00:00

BlockXXX is a new extension for Firefox that blocks pornographic content (the anti-Pornzilla, if you will). According to its release notes: This seems to work pretty well, although there are definitely some holes (large and gaping). Definitely not working well then.

Until about a month ago, all Firefoxen contained the legend Cookies are delicious delicacies. to describe cookies. Then Mike Connor decided that Cookies are pieces of information stored by web pages on your computer. They are used to remember login information and other data.. Distraught, I incited the creation of an extension to restore the legend.

A month passed and the world had all but forgotten about our delicious delicacies... until Jesse Ruderman published the Delicious Delicacies extension, dramatically subtitled Restore the legend. The world rested in peace once more.

Bye Bye, Sharpreader

2004-03-14T20:30:00+00:00

Since the demise of the free FeedDemon betas, I'd been using Sharpreader or Wildgrape NewsDesk (depending on what mood I was in) to read RSS feeds. Neither was perfect, but I'd settled on Sharpreader, despite the oh-so-slow notification pop-ups (this on a 2.5GHz machine...). But today I thought I'd give the RSS Reader Panel for Firefox (http://texturizer.net/firefox/extensions/#rssreaderpanel) another try. Guess what? It's pretty good. The Open In Contents Area option now produces a spiffy-looking display, with the option of customising the CSS used (so you can't complain even if you don't like it).

The addition of new feeds isn't as slick as it might be, but it's just as easy as creating a new bookmark ...because you just have to create a new bookmark. There doesn't seem to be an option for intermingling all of the items into a single date-sorted list, but that's not too much of a worry.

VirtuaWin

2004-03-12T17:45:00+00:00

Multiple-desktop-ery is allegedly a somewhat standard feature of Linux desktops; here on Windows, however, it's incredibly cool. Two desktop multifiers for Windows that I've recently used are DoubleDesktop (http://www.fatfreesoft.com/) and VirtuaWin (http://virtuawin.sf.net/).

DoubleDesktop, as the name aptly suggests, doubles your desktop's width; windows can span desktops and there's a nice, simple switching arrow icon in the system tray.

VirtuaWin doesn't allow windows to span desktop, but it does pretty much everything else. You can have up to nine desktops (or up to several tens of thousands if you fancy editing config files), in any rectangular arrangement you like. To switch desktops, you just move the mouse and keep going; or, if you prefer, you can require one of Alt, Ctrl and Shift to be pressed; you can specify a delay of up to four seconds to prevent accidental switchage. You can cycle through desktops, move to the next desktop in a given direction, and switch straight to a specific desktop, using configurable keypresses.

Other than the lack of desktop spanning (which is pretty fun), VirtuaWin's pretty damn good. And it's open source.

Winwall

2004-02-22T02:25:00+00:00

I had been using Microsoft's Wallpaper Powertoy to rotate my desktop backgrounds; I now use Winwall (http://net-session.com/winwall/). The biggest limitation of Microsoft's Powertoy is that it only recognises GIF (boo!), JPEG and BMP images - i.e. not PNGs. Winwall, however, recognises all four formats (and probably more), and its automatic resizing looks much nicer. The program is closed source, but freeware.

Take Back the Web

2004-02-17T19:23:00+00:00

Mozilla Firebird has reached version 0.8 and has been renamed Firefox; Mozilla has an FAQ explaining why. That gets the obligatory announcement out of the way; now on to the original content.

Still To Do

Ben Goodger's Firefox roadmap outlines what will actually be included in the 1.0 release; this is a collection of other noteworthy shortcomings (a.k.a. “pet bugs”).

Don't Steal Image Associations

Currently, when you set Firefox as the default browser (in Windows, at least), it automatically assigns itself as the default application for PNG, JPEG and GIF images. In the days when Microsoft Paint only handled BMP bitmaps, this sort of thing was OK; but now the ability to not only view, but also edit these types of images is built in to Windows, Firefox has no business associating itself with them.

Uninstall Extensions

Firefox won't be widely adopted by businesses and workplaces until one can easily remove any extensions that are installed. Bosses don't like their minions making any changes to their standardised computing environment, let alone irrevocable ones.

Single Window Mode

This one's a bit chewy. When its developers say Firefox is a “tabbed browser”, they mean it has the capacity to open multiple pages in one window. However, some folk interpret “tabbed browsing” as where all documents are opened in the same window - always; there have even been complaints that tabbed browsing is broken because this isn't the case.

Until Firefox does have the option for single window mode - which won't be until after version 1.0 - the developers should go easy on describing Firefox as a “tabbed browser”, to avoid disappointment.

Miscellanea

There are various other minor improvements that would easily and quickly make Firefox friendlier, and just plain better. The Windows installer should ask before creating Start Menu, Quick Launch and desktop shortcuts - it's only polite. The link toolbar present in Mozilla should be there in Firefox, too; perhaps then more people would start using <link>. And finally, MNG support. There's really no excuse for its absence - there's a patch waiting which only needs the thumbs up.

The Location Bar

The location bar is, in my opinion, Firefox's best feature, and the one thing that stops me from liking other browsers too much.

Any phrase typed into the location bar gets I'm-Feeling-Luckied, courtesy of Google UK, except for phrases which include dots. These are interpreted as URLs and invalid URLs generate an error page... but I can get around this using smart keywords.

Keywords

Firefox's bookmarks can be assigned keywords, which you then type into the location bar and - Hey, Presto! - the bookmark loads. This is great, but smart keywords are even greater.

If the URL of a bookmark contains %s, and the bookmark is assigned a keyword, anything you type into the location bar after the keyword (and a space), will replace the %s in the URL.

So, if I want to I'm Feeling Lucky some search terms (including ones with dots in them), I type goto followed by a space, and then the terms - simple. To enable this, all I had to do was create a bookmark whose URL is http://www.google.co.uk/search?q=%s&btnI=I'm+Feeling+Lucky and assign it the keyword goto. When I type, for example, goto hell into the location bar, Firefox converts this to http://www.google.co.uk/search?q=hell&btnI=I'm+Feeling+Lucky, which results in Google sending me to its first match for “hell”. The whole process takes less than a second.

I've also got other smart keywords set up for Google Image Search, Mozilla's Bugzilla database, Dictionary.com and IMDb.

All of this means I've practially done away with URLs. If I want the BBC's website, I just type BBC, and Firefox and Google do the rest. This ludicrous ease of use is, for me, Firefox's killer feature.

Google UK

By default, Firefox uses Google USA, but you can change this in about:config; it's pretty easy to do.

First, type about:config into Firefox's location bar and press Enter. You'll be presented with a plethora (or two) of settings. Into the box next to Filter:, toward the top of the window, type keyword and press Enter; this filters out the other settings we won't be using.

Double-click the line containing keyword.URL and enter http://www.google.co.uk/search?btnI=I'm+Feeling+Lucky&q= into the dialogue box that pops up. (I'd copy and paste it.) Finally, make sure keyword.enabled is set to true, and Bob's your uncle.

Brand New

As mentioned above, along with the new version came a new name and brand, Firefox. The previous plan was that “Mozilla Firebird” would be the project's code-name, and it would eventually be known simply as “Mozilla Browser”. Gladly, that's now changed, and we have a browser whose logo looks like this:

The new identity lays to rest previous dissent over the icon/logo Firebird was using - an image of red and orange flames, whose form was also reminiscent of a bird's feathers (this logo is still visible in Firefox 0.8's Help > About > Credits screen). While I always liked it, some felt it stood out too much from other Windows icons (which in my opinion is good), or that it became an amorphous red blob when shrunk to 16×16 pixels.

This name will stick - it's been thoroughly researched, and no-one else is using it for anything resembling a web browser. This means you can start posting your favourite Firefox propaganda about the web (and anywhere else), in good conscience that it'll still make sense in a few years' time. All of which is lovely, as the logo looks bloody brilliant!

Doing the Job

The purpose of a web browser is to display web pages. Mozilla Firefox uses the Gecko layout engine, which displays pages more properly than, say, Internet Explorer. By “more properly”, I mean Gecko better conforms to web standards, as described by the World Wide Web Consortium. Firefox displays web pages better than many other browsers - it does the job better.

Incidentally, Mooquackwooftweetmeow conforms to these web standards (as should all websites), with no regard for how Internet Explorer mangles its pages, so Mooquackwooftweetmeow looks better (i.e. decent) in Firefox.