The GIMP, Woes

by Kieran

Now that my university year is coming to a close there are many doors opening in the future of my software development. Fortunately I have a handful of people who are very eager to work together on solo-driven, yet team-oriented software projects. As a part of this we’re setting up our old team, and reviving old projects that have been left for dust to settle on them, and hopefully with enough time to get something decent developed. As part of all this setting up, I’m having to develop a unique website, as some of the projects will be for sale (although none of my own will be, only some that I will work on).

I’ve started to use GIMP instead of Adobe’s Photoshop for developing graphics for a number of reasons which I won’t go into now. I’m a big fan of Open Source Software (OSS), though occasionally I do have certain “issues” with their usability features. There are a few projects lurking around the Internet that are almost synonymous with that of commercially-released products, for example, Splashup being similar to Photoshop, and Canvas being similar to Microsoft’s Paint. Both of these are very good, especially on a usability level being extremely similar to that of the commercial counterparts that they mimic.

The problem that I find with GIMP is that its default behaviour is quite different, particularly in the areas of keyboard shortcuts. Now from the outset it is quite obvious that the user interface is different from that of commercial competition, yet I don’t see the need to switch around all of the keyboard shortcuts. I have seen scripts which allow an exact replica of Photoshop’s interface to be used in GIMP, but then many people may disagree with that.

My main issue - after all of this - is that some projects attempt to be different from big projects quite simply to “be different”. Some of the things I learnt in HCI theory was to mimic other products because it’s obvious that users will be familiar with their key usability. Sure, you can extend this to a certain degree, but completely changing every single shortcut just isn’t right. I sure won’t be doing that in some of the applications I’m working on.

In addition to this, however, I can’t see why the GIMP isn’t suitable for much development. It sure is going to take me a while to get used to all of the shortcuts, and I’m going to try to learn using the default ones as every time I reinstall it I won’t want to be applying scripts to the application just so I can use it.

Tarski is a very nifty and editable theme developed so much so that it has its own website. This is the second time I’ve had to reinstall WordPress on my hosting because of that theme eventually killing my administrator pages. I haven’t seen any web links mentioning that the latest version of WP and Tarski don’t agree with each other, and instead I’ve had to find out the hard way. Twice.

I’d recommend not using that theme for now, although it could just be something to do with my hosting. I’ve reverted to some other random theme for now, although I’m doing a bit of web development at the moment so I might just have a bash at making my own one. Not that themes matter too much, although it’s nice to be consistent (something which I’ve never learnt to do).

EDIT: Sorry, this was against WP 2.5.1, FYI.

Final Stretch

by Kieran

I handed in my research project yesterday which was just over twenty-thousand words, not that word count matters in these kinds of things but you can imagine how many hours of work that is.  In fact, OpenOffice tells me that I had done 350 hours of work on that particular document, which is more than I expected it to be.

I had about two hours sleep the night before hand in as we stayed up until 4am then got up again at 6am. It’s not that it was a rush job towards the end but we had read over each others (me and a friend that is) and found a large number of errors throughout each others documents which needed fixing.  Then there were just all of the loose ends such as appendices.

Printing and binding actually took about 3 hours, so it’s a good thing we got up so early to make sure we were prepared. We ended up handing them in at about 2pm I believe, so an hour shy of the deadline.

I’ve learnt a lot from the project, mainly time management, research skills, development skills and research writing skills.

Now I have two exams left which start in three weeks time, and I’m very much unprepared for them as I’ve been focusing on the final year project so much. I slept for 12 hours last night so today isn’t a good day to be studying as my body seems to be trying to want the caffeine that I’ve been feeding it four times a day for the last two weeks.

Oh also my WordPress install died, so the change of theme is because I’ve forgotten what the one I was using was called. I have no idea why the install stopped working, I hadn’t changed anything.

Music of the Month

by Kieran

As I sit behind a computer all day everyday it’s important to keep a focus on my work without any distractions.  Each individual works differently when sat behind a computer, I know some that prefer silence, and some that keep desktop toys to stop them from going mad (I’m talking of things you get those gadget sites).  My sanity-focus is music, so I tend to listen to a lot.

Recently I got a hold of Ani Difranco’s album "Canon".  I’ve been a fan of Difranco for a couple of years now and I have only ever owned her ever popular Knuckle Down album.  If you don’t know who Ani Difranco is, well, she’s an acoustic folk artist with a passion for singing about unspoken things such as sexuality, homophobia, drugs, money etc.  Her style of music is confusing to say the least with a heavy acoustic style and a passionate voice.

The album itself contains many of her older songs but "revamped" as so to speak with live versions also.  I must say that my favourite song is "Fuel" which is a very rhythmic track containing some extremely absurd, yet true statements.  It almost sounds as if she’s rapping, though it’s more of a rhyme.  I can’t quite put my finger on it.

I definitely recommend this album, and it’s rare that I recommend music.  Perhaps find some previews of tracks somewhere, or jump on Last.fm and try and get a listen on some of it.

I remember about two years ago I would proclaim my dislike for coffee to my housemates at the time.  "It’s foul stuff!" I would say without thinking twice of taking a quick sip to glimpse a recollection of the horrible taste.

For a period of time I was an absolute health freak, I wouldn’t even eat a chocolate bar or have a fizzy drink.  I went from a thirteen stone, five foot four, sixteen year old to a nine stone, five foot six, eighteen year old.  Previously I led a care-free lifestyle with absolutely no consideration for the foods and drinks I was consuming on a regular basis.

Of course there’s a picture of me of when I was at my skinniest (I’m on the right, picture taken 2004), and I find it mildly disturbing.  Three years later I weighed in three stone heavier.  My weight has been up and down over many years, and again I lost stones worth of weight during my stay in hospital and the time afterwards.

Though despite all of my health-concious years I still had never touched a cup of coffee and stood religiously next to my bottles of water.  Let’s take that and put it into perspective.  Yesterday I believe I had five cups of sugar-spiked, three-spooned-strong coffee and I loved every sip of it.  I first tried coffee when I started having Mocha’s at a company I used to work for (we had two Costa Coffee’s in our buildings).  I think I’ve been hooked ever since.

With a fiancee who regularly drinks coffee, and a software developer whom I work side-by-side with drinking it as if it were the cup of strength there is no way for me to avoid it any more than I’d like to.  The question is, does it really help?  I have actually been drinking caffeine-free coffee which tastes very similar, but the health risks of caffeine-free coffee are supposedly greater than that of regular coffee due to the chemicals used to remove the caffeine, plus, you don’t get the "hit" from it.

What I will say is this.  I’m giving up as soon as this research paper is due in, and I’m going to get quite health-concious because I believe I am doing far more harm than good.  Stay safe kids, avoid the dream-beans.

The Final Stretch

by Kieran

Well it’s Tuesday night and in one weeks time I will be taking my research project to a print shop to get it printed and bound.  At this point I’m loving the fact that I can use the word "I" as my project disallows this kind of language use.  I expected the project to reach about 12,000 words at the most, and I can safely say that there’s sections which will be at least 4,000 words that aren’t touched yet despite my word count now being almost 14,000.  Of course word count has absolutely no impact on the quality of work and projects with just 10,000 words can do better than projects with 20,000 words.

I’ve been trying to ignore the word count and am planning on cutting out any irrelevant text in the days leading up to the hand in.  This Saturday a full project review is being done to tie up any loose ends from a third party perspective.  It’s been a really interesting project I admit.

I don’t plan on it stopping there however.  Personally I think I’ll take it to the next level and re-develop the application from scratch and turning it into something usable by the general public.  It has also given me enough insight to hopefully give some talks at the GSDF on network-enabled application programming in the summer if they allow me.

Anyway, I need to be strict on my bed time and it’s already 10:30.

WordPress 2.5

by Kieran

Following Damien’s post on WordPress 2.5 I decided to upgrade from quite an old version that I had installed.  I must say, I am amazed at the amount of changes that have been made.  The AJAX-esque improvements are tasteful.  The features that were most significant to me are as follows:

  • Permalink shown immediately
  • Tagging on the fly
  • Improved widget control
  • Improved administration interface design
  • Advanced media inclusion

There’s loads more but those are just a few I noticed from briefly using it.  The only thing I dislike so far is how tag selection/categories are placed underneath the input box which means you have to scroll to complete a blog post which is quite frustrating as you didn’t have to do that before.

I’m sure there are many more improvements to be made in the future, and this is the beginning of far better versions of WordPress.

I love OpenOffice.org, I think the team have done a splendid job, and I particularly can’t wait until OpenOffice 3 is released as outlined by OONinja.  Before I begin I wanted to clarify that I have nothing against the team, or the work they’re doing, just in case this seems like a true rant.

I’m a sucker for optimisation, and I particularly like keeping my "Temp" folder clear in "C:\Documents and Settings\Kezzer\Local Settings\Temp".  It gets full quite quickly with arbitrary files that serve no particular use.  I clear it out every few days and sometimes find 20MB > worth of temp files that serve no use.  At least, that’s what I thought.

Now onto OpenOffice Writer.  I’ve recently been getting error messages crop up saying "Error writing file" when attempting to save, or auto-save.  I would have edited a large portion of my document only to find I couldn’t save it.  I promptly copied the entire document and pasted it into a newly created one and was able to save it.  However, I had never put temporary files and writing file errors together, I had never seen the link and of course it may be hours until my document was saved after clearing the temp folder.

There’s a folder that exists in there (for me anyway) called "svhc4.tmp", if this folder is deleted then the relative OpenOffice session loses its ability to save the current file.  I assume the folder name is relative to the current session and perhaps is incremental.

Let’s be straight here - I’m not a fan.  I know there’s quite a few facebook haters in the software community (understandably so in some respects I must admit) so this will add fuel to the fire.  One great feature of facebook is the ability to create events, detailing everything about the event and then inviting members.  Myself and Amanda decided to create one for our wedding, quite a few of our friends have already done it so we thought it would be a good idea to spread the word in a quick and easy manner.  Well, I thought it would be quick and easy.

Sure, facebook has some quirks, but this one is particularly frustrating.  When adding members to an event, you have to add them one by one.  That’s right, I’ve got over three hundred friends on my facebook account.  It gets even worse.  You can only add one hundred at a time.  That means I had to click each and every individual, but I couldn’t do it in one go.  It’s fancied up to the eyebrows with AJAX enabling you to select each friend and they’re nicely added to the left pane, but it’s a painstakingly slow process.

Dear twenty-something-year-old-zubrick-dude,

Please add an "invite all members on my friends list" button to the event management page.

Kind regards

An annoyed (yet using your product for free) user.

I guess not everything can be perfect straight away. Oh well.  Why they limit the number of invitations to one hundred at a time is beyond me, perhaps their systems aren’t designed to deal with that? I have no idea, but as an end-user, I shouldn’t need to know.  End-users should ‘expect’ things to work, despite their free-use policy ;)

I’ve had my fair share of software experiences.  Sure, they may not be real life experiences and they do sound particularly ‘geekish’, yet the use of software in some cases relates to the use of physical objects in the real world.

Recently I’ve been writing up my research paper on a network-shared whiteboard.  Whilst working on this I’ve also been looking around at various software as well as using various new websites.  I won’t mention these on here as so to avoid any flaming, but I’m sure people would tend to agree with my experiences.  The problem is - you see - that software, and their interfaces are more complex than they need to be.

One website I’ve recently taken an interest in is currently in beta.  A few days ago I checked it out to see how it’s going and sure enough on a foot-note there was a few small statements telling users what to do when they encounter an error, and how to interpret them.  Here’s a few quick facts for you.

  • Users don’t care.
  • Users just want usability.
  • The simpler it is, the more comfortable users will be.
  • The users probably wouldn’t follow up errors anyway.

During the development of my application I have made it one of my primary aims to hide as much as humanly possible from the end-user.  If I imagine myself as using an application I typically want something that "just works".  I don’t want to have to set anything up, I don’t want to have to go through complex instructions or be faced with indecipherable error messages.

Of course we have to take into consideration the end-user themselves and what type of user they are.  Some end-users may be a developer and therefore aware of how to deal with such errors.  For applications which I personally develop, they’re aimed at a very large user base, they may be developers, students, people in finances, or y’mum.

I can’t help but feel deep, technical hate for over complicated applications driving users away.  Some of the most popular applications in the world remain so utterly simple that a five year-old could (and indeed do) use them.  Google, for example, have a simple, intuitive interface keeping style and colour consistency throughout.  Have you ever used their calendar system, then moved onto their forum system followed by their mail system? You’ll find if you can use one, then you can use many others.  I am not, however, trying to "big up" Google in this post, I’m just using them as a prime example to the KISS statement.

So, here’s a few tips which I’ve managed to learn from, and acquire over the years.

  • Error messages should be on the level of a user
  • Error messages should only be displayed if a user needs to know
  • Categorise your layouts, both menu-driven and general GUI’s
  • Keep a visual consistency throughout your application (I’m talking about dialogues, messages, menus, option panels etc.)
  • Remove unnecessary technicalities (one-click installs are an example, but provide advanced features only if needed)

I could go on forever, but those are some of the prime examples.

I’ve made a mention to this for one other reason, and that is OpenOffice.org’s bibliography feature (this is in version 2.0.4).  It appears to me quite complex, and something I just can’t get my head around.  The table of contents feature is perfect - it’s simple to use, and easy to maintain.  Trying to maintain a bibliography should be something of great ease, perhaps selecting an area of text and clicking "apply bibliographic reference, and please reference it at the end of my document in a reserved section".  I would’ve hoped the rest would be done for me.  A friend of mine then showed me how Microsoft’s Word bibliography feature worked, and how stupidly simple it was.  I did have a deep envy for that feature in particular, although I wouldn’t turn to proprietary formats.

Wordpress itself, that is what I’m writing in at the moment, has the two-step install feature - what more do you need to specify?

So, to sum up, I personally feel that as developers we see software from a developer perspective.  The features we look for lay not in simplicity, but choice.  We lose sight of whom we are directing our software at.  The next time you’re designing something, try to cut down on things you think really aren’t needed.

« Previous PageNext Page »