3 Things That Ubuntu One Should Sync, That It Doesn’t

28 Dec

Ubuntu One a syncing service that Canonical bundles with Ubuntu. I think it’s great that it’s bundled with the OS, but it doesn’t feel like it’s being extended to its full potential (or even most of it). Recognize that I’m talking about the syncing service alone, as opposed to syncing+music store+[insert new feature here]. As of right now, it seems just to be Dropbox++. Since it’s bundled with the system, there are a few life-saving ways that it could be integrated into the system as well:

1. Desktop Wallpaper, Icons & Panels

This one’s as basic as it sounds: just sync the Desktop folder, the Desktop background image and the panel set-up. The user starts up their brand-spanking-new Ubuntu system and finds their familiar set-up ready to go. No looking for a pretty desktop background image, no re-creating application icons on the desktop, no fiddling with the panels to get them just right: they’ve done it all once, now sync it so it doesn’t have to be done ever again.

2. Installed Applications

Put this in the Software Centre, please. Automatically record and sync installed applications to Ubuntu One, then install them all on your laptop and/or that new computer you just bought. Obviously, the actual installation shouldn’t be automatic: I don’t want all my desktop apps on my netbook or work computer, for example, but make it easy please.

3. Online Accounts & Wallets

Ubuntu’s recent push towards social networking and the like is great, as is their indicator applet, and Gnome’s built-in password storage implementation is fantastic for when I only want to remember one password, but it’s for naught if I have to set it up on every computer I need them on. Sync E-mail accounts, Broadcast accounts, Instant Messaging accounts and the login wallet for passwords so I don’t have to go through these set-up steps several times.

There you go: 3 things (well, technically 6 (or 8 depending on how you’re counting)) that Ubuntu One could (and maybe should) sync to make it a totally kick-ass service, compiled in a teeny-tiny blog-post.

Ubuntu, Rolling Releases and Support

24 Nov

There’s been a lot of chatter about Canonical moving Ubuntu from the current 6-month release cycle to a rolling-release style of distribution. First, I’ll just explain what the two distribution methods are, their strengths and weaknesses, and finally how Canonical fits into the picture.

Fixed Release Cycles

Fixed release cycles date waay back to the 1980s some time. Maybe. I’m not sure. All I know is, this distribution method is really really old. It’s tried and true. I’m going to demonstrate what goes on in this release cycle.

Start ——————————–> Code
This line represents time. As time passes, code gets added, shifted and taken away from the central project (in this case Ubuntu): it could come from new application versions (as with Firefox) or from Canonical (as with Ubuntu One or the Me Menu). All that’s important right now is to realise that code is constantly being changed in the project.

Start ———————————> Code
                              ` Stable release marked!

Now, the line has a branch coming out of it. Canonical have said, "These are the applications that we want to include in [10.10]". Now, the amount of code allowed into the project is being limited to fixing stuff.

Start ———————————> Code
                              `————–> Stable Release Released!

So Canonical have been working on that branch since it was marked, fixing stuff that people reported as broken. In the meantime, people are also adding whatever code they like to the project (whether directly to Ubuntu or indirectly through a 3rd part project like Firefox). The stable release is out and people are downloading it. Yay! Now, no new versions of applications make their way into the stable release and the amount of code going in is severly restricted to fixing important bugs.

Start ————————————————————–> Code
                              `————–> [10.10] — — — — — –>

So in actuality, when Canonical works on Ubuntu, they’re actually working on two (well, three, as explained later) projects – the current stable release of Ubuntu (dashed here), fixing things that are broken or vulnerable, and the next release of Ubuntu, which gets new features, new artwork and more sexy. The top line will eventually branch again into the next release [11.04] and when it does, support will drop for the current version. The exception is in LTS releases, that also get their own branch – this would go below the dashed line and carry on until every 4th release, when a new one comes out. So, Canonical actually keeps 3 code paths for Ubuntu: one is for lots of features; one is for bug-fixes for the next release; and one is a 2-year-long support code path.

Rolling Release

Rolling Release distributions are much more hectic to keep track of and aren’t as simple as the Fixed Release Cycle above. Rolling Release distributions let you install a, "base" system, then receive updates for applications progressively. Some distributions (such as Gentoo) let you install the very latest and greatest releases of applications when the applications get updated, though others can potentially hold back appplications until they’ve been proved to work with the system to maintain stability. Unlike the fixed release cycle above, applications will receive updates that include new features and new application versions. They receive much less QA testing from the distributors, but if the independent projects do their own extensive QA testing this may not be an issue.

Canonical: Rolling Release Vs. Tried and True

The question of whether Canonical should move to Rolling Releases is a question of cost-benefit analysis. I’m a tinkerer and a geek, most of my family are and I’m sure, if you’re reading this, you probably are too. There are also people that can quickly figure out what’s changed, how it’s changed and how they adapt to the change when new versions of applications come out (these people tend to be our younger generations, as they have grown up around and lived with computers all their lives). For these people, Rolling Release is very advantageous: we get newer software with newer features faster. When a new release of Firefox comes out, I want it – there and then, not with the next release of Ubuntu. When a new version of Rhythmbox, or Empathy, or Evolution comes out, or a new release of Amarok, KDE SC or KDEPIM comes out, I want to be able to install and use it without worrying about PPAs and without worrying about stability (except, maybe, in KDE SC – depending on who you ask). This is how third-party applications on Windows work, badly designed as updating software is on Windows, and rolling release seems to work well for consumers there.

However, in corporations especially, this isn’t OK for various reasons. First, keeping the software at a specific version for a long period of time (2 years for LTSes) means that, when there’s a problem, the guys in I.T know what to do to solve it. If application versions keep bumping every few weeks, with new features, changing layouts and the like, I.T could bump against a problem they’ve never experienced before, or experience an old problem only to find that the method of solving it has changed (menus, configuration options have changed etc.). This is a massive inconvenience. Then consider that many corporations may have specialised or in-house-developed software, built against only one version of a set of libraries underneath: if these keep updating and changing, or even if something as simple as the default theme keeps changing, it could break something in that application. Since we know that application is made in-house, or a software house has been sponsored to develop it specifically for them, we can assume that it’s pretty damn important and that having it break every few weeks could mean plenty of hours lost constantly trying to fix it. Not cool.

These aren’t the worst of it, though. Gather ’round this campfire, ’cause I’m going to tell you a horror story my father told me: he’s every bit a geek that I am and loves his technology, so he knows how to look after himself in the digital world. But at his workplace, in a hospitle, he goes towards the locker room every morning and evening which is locked with a key-code – except this night, when he approached that locker room door, he found on the wall beside it – not blood, but something much, much worse… the code to the locker room, written in permanent black ink. Terrified, he fled at the site, retreating to his familiar and comfortable computer desk, with the shared computer – only to find that there, too, was written someone’s user-name and password, on a sticky note on the side of the computer screen.

Horrific, I know – but the sad reality is that these people are a majority. These are the kind of people I.T has to cope with, and they seem to panic as soon as anybody suggests they move out of the groves they’ve worn in their minds. So, you can imagine, having the software they use change every week or so is a nightmarish hellride – no, not for the stupid plebs, but for the poor, poor guys in I.T. Think of the I.T department: offer a fixed release cycle.

All that being said, I don’t see any reason why Canonical can’t use both methods – simply take a, "snapshot" of the package versions every two years and slap on an, "LTS" sticker, then have everybody else use the rolling-release cycle. I certainly think it would be better for Ubuntu’s current user-base. I’m holding out hope, anyway. Well, there go all my hopes and dreams. :/

Question: Why Use a Blogging Client?

23 Nov

So someone very honestly asked, "Why bother using a blogging client at all?". Well, what better way to answer than in a blog post? Since I’ve just finished moving house and don’t get broadband installed until Friday (using my mobile phone), I figure this is as good an excuse as any to get back on the ball with my blog. Lets get to it, then!

First: Itty-Bitty Advantages

To start, I’ll just group up a bunch of 1-sentence, less-than-140-character advantages. Saving blog-posts locally is a life-saver for me on my netbook, for when I’m on-the-go and need to punch out a blog post. With Blogilo, you can also download your blog’s theme locally to preview your blog posts without an internet connect (again, an awesome feature for when you’re out-and-about). Using a client also means you don’t have to open your web-browser, go to your blog’s home-page or admin-page, then tell it you want to write a new blog-post – open the client and you’re ready to type, straight away. If you run multiple blogs, you can also just set them both up in the client, write out your blog post and pick which blog to upload to (or post to both). If you have different blogs with e.g. different blogging providers (WordPress and Blogger, for example), then it doesn’t matter to the client: it looks and works exactly the same way.

Next: Long-Winded Explanations

Screen-Space Stress Test

A simple experiment can be done to see which application uses screen-space most efficiently. In fact, it’s so simple I can explain the whole process in one screenshot:

Oooh dear. That website doesn’t seem to be coping well.

Yeouch. Hell, I’ve done everything I could to get Firefox to save space, and yet… ouch ouch ouch. Blogilo wins this one, even before you remove the toolbox.

The reasons are obvious. WordPress.com suffers all the weaknesses of trying to make an application out of a document-layouting language, while Blogilo benefits from the full power of modern-day desktop application toolkits: this is evidenced by the fact that, despite not having enough room for the Visual Editor tab’s toolbar, the toolkit knows how to properly manage that situation (hide some buttons at the end). Moreover, because you only ever open Blogilo to actually post something to your blog, that’s all it has to display: web interfaces also need tools and navigation for visiting your blog, managing comments, visiting other people’s blogs, managing your account, managing media etcetera, and that stuff scoffs space. The website looks, feels and works like… you guessed it – a website. On the other side of the fence, the desktop client looks, feels and works more like a document creator, which is good – since the focus is on the document and not the rest of the interface.

Lots of people complain that screen-space isn’t really that important, especially when you’ve got a large monitor. They’re half-right: screen space doesn’t matter if you’re only ever doing one thing at a time. However, the more screen space my blogging client takes up, the less I have for, say, a website I’m referencing, an E-mail I’m reading or what-have-yee. I like being able to see all this stuff at once, and the less screen-space that’s wasted on the application itself the better. If I only ever wanted to see one app at a time, I’d buy an iPad. Nooo, thank-you.

Features, Features, Features

That’s all well and good, but it counts for naught if the client just makes me do more work to make my totally awesome blog-post… totally awesome. Fun times, people! The client wins on this side, too.

Because the rest of the chrome takes up so little space, they can afford to cram in buttons that add extra formatting options (headers, lists, the standard BIU stuff, changing the text colour, alignment and tons more fun stuff). This is all stuff that (at least, in WordPress) I’d have to go into the HTML editor to enter in manually – which is annoying, especially for lists.

Managing multiple blog posts at the same time is easier, too – for two blog posts, you’ll have two tabs in your blogging client and nothing else. In a web-browser, those two tabs could be mixed in with any number of other tabs or grouped with some task in the task-bar somewhere, which can make it a pain to find them (of course, you can do stuff to negate those issues like just having your actual blog posts in a separate window from your main task, but with a blogging client I don’t have to do that kind of stuff). All the more comprehensive stuff is there as well – the current post’s permalink (if it’s already published), post summaries, scheduling posts, submitting posts as drafts, downloading current posts, submitting edits to those posts and deleting posts. I’m sure it’s something to do with a lack of relation to blogs, rather than technical feasibility, but the only thing I’m really missing is a, "make me a cup of tea" button.

But Most Important of All…

The client is prettier. :3

On a more serious note, getting a blogging client written in your preferred toolkit means you get a client that integrates well with your DE (and you can even make KDE apps adopt the GTK+ theme, giving you no excuse to use anything other than Blogilo 😉 ). That means it works and behaves in a way that you’re familiar with and in a way that you have already told the computer you want applications to behave in.

That’s all I’ve really thought of right now, but I hope it’s been enlightening. Thanks for reading, people. 🙂

Gnome Shell Review

14 Oct

OK, you all probably know about my unadulterated bias towards KDE, but I’ll keep that out of here. From henceforth, I shall not mention that other desktop environment during this review, neither for comparison of features nor stability. I’ll also kindly omit the presence of bugs in this review – there are some, but it’s beta software, and as I’ve stated before bugs are unintentional side-effects of the developers having souls, while design decisions are intentional and indicative of the direction of the project. So, I’m going to keep the discussion strictly to features and intended work-flow. Right, with all that out of the way, here we go.

What’s Stayed The Same

Most things, actually. Nautilus still draws the desktop, the same themes are used for windows and widgets, and all the programs remain the same. Whether there are massive changes on the way for the rest of Gnome as part of 3.0, I don’t know, but it should be noted that Gnome-shell will only replace your window manager and panel manager. Splended.

The Defaults

By default, you’ve got one panel at the top, which shows in this order: the Activities button, where all the interesting stuff happens; the current window’s title; the clock; the system tray (which hasn’t been ported to the new spec, as used in… erm, Ubuntu); and finally, a menu with your username on it. I don’t like that the clock has been placed smack-bang in the middle of the panel, with yarns of space on either side – it seems isolate and out-of-place. I feel like it should at least be grouped with the other applets, on either side of the panel. Clicking on it brings up the calendar which, being small as it is, only draws attention to the fact that it’s a teeny-tiny bit of information right in the middle.

The default screen

Now, onto the first feature (or lack thereof): there’s no way to change the panel at all. No right-click menu, no right-, middle- or left-click-‘n’-drag of applets, no adding or removing panels – nothing. You get what you’re given and you’ll like it, dammit!

Well, I don’t. Like I said, I don’t like that the clock is smack-bang in the middle. I’m not asking for the world here: I mean, I would be fine if I couldn’t make new panels, resize the current panel or even change where on the screen it is. All I want is to move that clock. Surely a click-‘n’-drag should be enough. I’d prefer it on the right, as I’ve always had it, but at this moment in time that requires that I hack the javascript that the panel’s written in (or something daft like that). Haha! Thanks but no thanks.

Now, this may well come in the future, but that it isn’t here already is pretty disconcerting: I know not everyone is the tweak-geek that I am, but that even the most basic amount of customisability seems a complete afterthought is quite distressing.

Anyway, as shown in the screenshot above, the theme is currently white-text-black-background. Fair enough, lets focus on getting the functionality in there and getting it working to start. However, there are actually themes available for it and, as a natural progression from that, I expect the ability to change the themes to be built-in with a nice GUI somewhere. Doing a bit of googling lead to this post, which does list some quite attractive themes – I couldn’t get one of them to work, but no biggy. I’ll try again later and make amends to this post if warrented.

What You’re Here For: Workflow

OK, so to start I’ll cover the more basic right-hand-side menu. You can go here to change your instant messenger status, to change your account or system preferences, to toggle a (currently ugly-as-sin) side-bar and to log out. It seems to me to be a strange kind of menu, with several seemingly disconnected activities going on in there (for example, I expected preferences to be part of the Activities menu), but there’s not so much in there that you forget what to go there for.

Now then… the Activities menu. The meat-and-‘taters feature of Gnome shell. First things first, I’ll show you what it looks like:

The display you’ve all seen before

There’s some pretty cool stuff here: you can mouse-wheel over windows to, "zoom in" on it until it’s the actual window size; click-‘n’-drag applications, documents and places onto a virtual desktop to open it in a window in that desktop; see all windows across all desktops and quickly switch to any of them, or close them, from this view; create and remove virtual desktops; and search for applications, documents or places to open.

I don’t like it. I’m sorry, I really don’t. It’s nothing to do with the theme, either, I just don’t like the way it works. I’m stirring up a hornet’s nest with this blog post, but I just really, really don’t like it.

First, why does this assume that I want to see everything when all I need is, say, to close my Thunderbird window? To open my home folder? To switch to my Firefox window? Why do I have to move my mouse to the top-left before I can switch back to my blog post window, just to move my mouse back to the bottom right? I get that virtual desktops needed to be made more obvious, in a way, "explained" to the user through the interface itself, but why should I need the ability to add and remove desktops when all I want is to open OpenOffice? Am I really going to be adding and removing virtual desktops often enough that it warrants going there? Isn’t there a more economical way to either launch applications or manage virtual desktops?

And why, O why, does the application menu not divide into categories as in previous versions of Gnome? You know why there wasn’t a search function in the previous menu implementations, right? Because they didn’t necessitate one. If you’re going to add a search function, make it complement the current, very nice menu structure, not replace it. Having all applications everywhere exposed all at once, in alphabetical order, is just as bad as the Windows start menu. Seriously, this menu is horrific. If my hand is already on the mouse, say, to move to the menu, then I’d like to be able to open, say, Thunderbird with one swift motion, not having to scroll through hundreds of applications to find it. Yes, I can hit Alt+F1 and search for Thunderbird, but maybe I don’t want to because maybe previous versions didn’t necessitate that behaviour to get decent results in a sensible time frame.

Why isn’t there another, more sane way of accessing minimised applications, than the Activities menu? Why is my list of open applications mixed with my list of favourite applications, and how do I tell the difference when I have, say, my favourite applications open? Why does right-click behave differently between the open/favourite applications and pretty much every other menu in there?

How am I supposed to be able to tell, at a glance, my applications apart from the Activities menu if I have more than 6 windows open? Bear in mind that this is on a huge screen: what am I going to do if I try to use this on a netbook? Doesn’t this lack of recognition defeat the entire purpose of laying your windows out on a grid Compiz-style? Why can’t I move that sodding clock?!

…*cough* Suffice to say, it’s progressed well, but is currently still quite rough around the edges, and for the first time I think I can say I’m excited about Gnome (Jesus, did I just write that?). I’m looking forward to the final 3.0 release, when hopefully the bugs are ironed out and the features are figured out and implemented, so keep up the good work!

Latest Firefox Improvements: Improved Location Bar, Panorama Tabs And More

29 Sep

So there have been quite a few notable improvements in the very latest version of Firefox (meaning the daily build, called Minefield) that I think you people will be quite interested in. First, I won’t cover stuff I’ve already covered in my previous review, so go check it out first if you haven’t already. Now, lets dig in.

New Location Bar Improvements

The Location Bar has seen some new features. First and most notable, the Stop/Reload button(s) have been removed from the toolbar and moved into one, combined button at the far right of the location bar: now, it changes between Go, Stop and Reload depending on the state of the current tab. Nifty.




If you’re panicking or annoyed at the sight of it, don’t worry: it’s not hard-coded. What do you think this is, Google Chrome?

From here, it’s a simple click-and-drag. Simple.

The more observant of you will have noticed a very pretty blue glow in the second image: say hello to the new loading bar! Instead of a loading wheel in the tab, we now have a loading bar in the location bar for the current tab, or at the top of the tab for other tabs.

New loading bars. Functional and beautiful.

Additionally, hovering over links now shows their destination in the location bar in a bread-crumb-esque presentation:

Go-to locations in the location bar: heralding the death of the Status Bar, perhaps?

The clever thing is, if the whole URL can’t fit after the current URL, it squeezes the current URL down. If it still can’t fit, it makes sure to show the entire server, abbreviates the stuff after it, then shows the end of the filename as well as the extension:

Shows the stuff I’m most interested in, for security reasons: where am I going and what kind of file is it?

This is a really clever little extension to the location bar and, though you can still show the status bar (because lets face it, nothing just "dies" in Firefox), it’s no longer absolutely necessary if you want to stay secure. I always thought it looked tacky, anyway.

I think that’s all the location bar changes taken care of. Next up: Panorama!

Panorama Tab Management

Panorama is a new tab management implementation especially designed for the power-user. You know, the ones that love Firefox because it runs smoothly with more than 40 tabs open at once. It lets you zoom out to see an overview of all your tabs, organise them into groups and close entire groups of tabs with one click.

The zoomed-out interface

Using groups. Yes, still unemployed 😦

In case that wasn’t good enough, there’s also a search feature: just type and go!

Huoh. I wondered where that tab went…

Even though I rarely use it because I only ever have a few tabs open, I still think this is really really cool. Firefox is still your friend if you’re a tab-junky.

"And More…"

All right, I’ll admit, I didn’t plan this section through. "And More" sounds good at the end of a title.

No, wait! Don’t go! I found something to go here. Honest.

First: Google Chrome’s Pin Tabs.

Pin through the right-click menu…

Pinned tabs will stick around after Firefox restart, and won’t scroll when the tabbar fills

The Add-ons tab has been much improved:

The Add-ons tab now looks the part in Firefox

And the, "Get Add-ons" page has been much improved as well.

The improved Get Add-ons page – now more attractive and useful

Well, that’s my overview for right now – if more changes come this way before 4.0 stable, I’ll make sure to post about them.

Powered by Blogilo

OK, Seriously, Shut Up About Android’s Fragmentation

19 Sep

So I very often see arguments about Android and it’s, "fragmented"/"diverse"/whatever nature and whether it’s a good thing or a bad thing. Very often, it’s iPhone users that are complaining that first, Android is fragmented and that’s bad for consumers, and second, that fragmentation makes it harder for developers to write applications for Android. Neither of these make any sense to me what-so-ever, and I’m going to explain why.

Fragmentation: The Consumer

This is a rather silly argument to start with, because we already have another very real, yet very successful case of, "fragmentation": walk into any computer store and look around. There are computers running Windows Vista and Windows 7, and sometimes you’ll even see a netbook running Windows XP. If fragmentation is the overwhelming issue that I’ve heard time and time again people say it is, then computer users are in some nightmarish hell because they want Windows 7 and, horror of horrors, there’s a Windows XP machine right there!

The alternative: the user walks past the Windows XP machine and buys a Windows 7 machine.

In all seriousness, when did this become a massive user-facing issue? If a user doesn’t want an old Android 1.6 device, they can walk past it and buy an Android 2.1/2.2 device instead. If they are content with Android 1.6 (as many people are content with Windows XP), they can instead opt to buy an Android 1.6 device. There is no, "horror of horrors" because Android 1.6 and 2.1/2.2 devices are sitting on the same shelf.

People also like to complain that the same applications won’t run on 1.6 and 2.1/2.2. Quite frankly, this is only half-true: applications written for Android 2.0+ won’t run (fully) on 1.6, while applications written for 1.6 will work with 2.0+. This is no different to running a Windows XP application on Windows 7: sure, you might not get pretty Aero glass, but for a good application it won’t matter anyway.

Fragmentation: The Developers

Non-developers also like to complain that developers will have a hard time with all the different possible screen resolutions and versions of Android, which is hilarious in itself. Anyway, screen resolution is a non-issue: when you write an application, you’ll likely write it as a series of widgets (buttons, content views, text boxes and the like) that you lay out in a flexible grid. That’s why Google states that Android applications written under the Google UI design format will simply rescale to any resolution.

Now, the main issue can occur when coding for different versions of Android, but this is just an issue of whether you want to hit the largest user base or use the most new features. Windows XP is still the most widely used version of Windows, and will be until people need to replace their computers, so you’ll likely sacrifice pretty Aero glass for the ability to write one application and run it on Windows XP, Windows Vista and Windows 7 without worry. Similarly, unless you absolutely need some major new features in Android 2.1 or 2.2, you can likely write your app for 1.6 and hit every Android device out there. Alternatively, if you are so bothered about it and can be bothered to do something about it, you can write two versions of the app, but I don’t think there’s anything massively new and important in Android 2.0+ that you can justify that choice.

So, all things considered, I don’t think fragmentation will exactly kill Android right now, and with Google moving to yearly releases I think it’s going to become less and less of an issue.

For crying out loud, not another damned box!

2 Sep

So Apple just recently released the new iOS-based T.V set-top-box thing, Google’s going to release their Android-based Google T.V system (with Logitech providing a set-top-box on launch) and Reddit’s bombarding me with adverts for the TiVo internet-enabled… set-top-box.

Look, I’ll just tell you how it is, plain and simple: I have a T.V with remote; a Sky box with remote; a DVD player with remote; a surround-sound system with remote; and finally, an X-Box. My T.V rack is kinda cluttered, as is my remote control holder. What makes all you people think that exactly the kind of thing I want to blow money on is yet another God-forsaken box with yet another God-forsaken remote?!

Look, I’ll tell you what the clever thing to do would be: make a T.V. Include the basics: wide-screen plasma, SCART sockets and the like. Include Freeview built-in. One box, one remote down (technically, this doesn’t help with the Sky box any, but I’m going lowest-common-denominator here). Now, put a CD/DVD slot in the top and have playback controls on the T.V’s remote. There. Box/Remote combo number 2, down. Now, add a wireless card and let it connect to a LAN: this can be set up in the T.V’s settings or something. Write a piece of software for PCs(/Macs/whatever) that lets you click-n-drag files and websites to an icon and have them play on the T.V somehow. Yet another box, yet another remote, thrown out. Slap a logo on it and sell it as the next revolution in T.V technology. Job’s a good’un.

Seriously, if you need someone to come up with this stuff for you (and manage specifics etc.), I’m unemployed. Send me an interview request if your name’s Google.