Will Knott

Every so often you discover that some programmers are cool.

I don’t mean in the really skilled (can write a million euro application in 10 lines of code) but actually cool.

For example Luis be Bethencourt Guimerá one of the presenters at OSSBarcamp. No only is he a talented coder and software designer (who worked on the forthcoming Ubuntu release), but he also wrote his own VJ software (being released in the Jaunty Jackalope release of Ubuntu Studio) and digital DJ software, complete with intuitive interface, to help his gigging around the world.

That is cool! I’ll cover more on this later.

This OSS Barcamp was less like most other Barcamps I’ve been to in that the schedule was locked down in advance. It made an interesting change, but it didn’t stave off everyone’s technical difficulties.

OSS Barcamp has its worthy side too. Éibhear Ó hAnluain took the opportunity of a number of skilled and focused minds to look at “How to present a political party’s FLOSS-friendly IT policy to the electorate” or rather, what are the benefits to the country for using free and open source software. What I loved about this talk was actually Enrico Zini and the impact that open source software had on the Italian political scene and its civil service.

The lightening talks were fun, I missed part of James Larkin’s “Intro to CSS Frameworks”, but there should be a video of his talk (and I’m sure he’ll release his slides). One other thing which stood out is the effort to translate or localize Ubuntu to Irish. If you are a native speaker, talk to them.

I’m also fascinated by the formation of TÓG, hackerspaces in Dublin. In the same way that co-working has benefits, I can see similar benefits with co-hacking. And folks, this is the old style of hacking as in making things work, work beter, or creating something new out of other products. Not cracking which is breaking in to things. Think more of an organised “voiding your warranty”.

Due to the lighting talks I ended up missing David Coallier’s “Get Ready for web 3.0 talk”

Now, back to Luis.

First off watch this video.

Some RSS readers may need to click through to see the video

Looks pretty good. Not too fancy and suits the song.

Now just think that this video was created with

• one $160.00 disposable pocket video camera
• one batch of Free Software, most of which is in Studio Ubuntu
• two hours of shooting
• four hours of production

Cheap? And good looking too.

And then there is “Big Buck Bunny” created by the open source Blender tool. Now it took a long time to put together, but the Peach Movie Project Team include tutorials on how they did it.

Just remember kids; software is an art as well as a science. And artist in one field tends to have skills in other fileds.

Take care, and enjoy the short movie below,

Some RSS readers may need to click through to see the video

Thanks to Laura Czajkowski and Jaime Hemmett for putting the whole day together.

Will Knott

If you are looking for me tomorrow, I’ll be at the OSS Bar Camp in Kevin Street, Dublin (with my badge bag, makes me easy to spot I suppose)

Today is Ada Lovelace Day. Miss Lovelace,was the world’s first computer programmer, and the day is to promote women in technology.

Oddly for me, women have had a lot to do with my personal computing history.

I grew up with the low level grown and high pitched squeal of arcade machines. Computer games. So when the chance came to learn how to program a game on the old Commodore 64 I leapt at the chance.

The teacher in the classroom teaching Commodore basic, was a woman (whose name I simply cannot remember. Kineally or Keneally. Back then every teacher was either “Miss” or “Sir”). At this point I’m not going to name names unless she blogs, or at least tweets.

Later on, I watched the “video” games review TV show (or rather several, but all) hosted by Aleks Krotoski who still has a hand in the games world and quite possibly will never leave technology in particular how it impacts on people.

Later on came college (if the class wasn’t an odd number 50% of the class would have been women) and work in general.

Indirectly I worked under Padmasree Warrior, and female programmers were no different to the male programmers in the eyes of the code in Motorola.

Later the world of social media introduced me to Ellybabes, the first person I met at the first Irish Barcamp in Cork. To the coding evangelist (which might actually be part of her job title) that is Martha Rotter. To first Irish podcaster of either sex I met, and business guru Krishna De. To the promotional expert that is Maryrose Lyons. And to the best, and probably most fun, web designer; Sabrina Dent. To women who actually get thing done, like Laura Czajkowski and Alexia Golez.

So why isn’t there equality among the sexes for a job that normally requires communication, concentration and heavy thinking? It could be fictional role models and expectations as Naomi Alderton suggests.

Maybe.

But if you have a daughter, let her know that computers aren’t just for the boys. And if you are involved in technology yourself, take a look at the Geek Girl Dinners.

take care,

Will Knott

Related articles by Zemanta

• Pledge Ada Lovelace Day (emilychang.com)
• UnConferences In Ireland, March 2009: Dates For Your Diary (krishnade.com)

Hello again.

This post is going to be very image intensive, and I’m used thumbnails to help the bandwidth impaired.
Click on the image to see a larger version.

A quick word on Selective Colourisation, someone asked “what is it”…
Well, this is a colour image.

and this is the same image using selective colourisation on the yellow parts.

Effective isn’t it. The technique works best on a strong colour, unfortunately pink has many shades. Something which complicates things later on. And when yo do this technique on your own images, clean them up first. Red eye looks silly in black and white. However you could try selective colourisation and have a black and white photo, with red glowing eyes. Very zombie.

Anyway, in part one you downloaded your tools, and I gave you an image to play with.

So lets start.

Open the photo and you’ll probably have this view. The picture itself and the toolbar. However, in order to do make the changes, you’ll need to see the layers. To bring up the “Layers Dialog window” from the menu bar of the photograph, choose “Dialog” and then “Layers”.

And you then have the layers window.

What we are going to do first is create a new layer (a copy of the photograph) and turn that copy in to a monochrome version.

So go to the menu bar of the photograph again, and this time choose “Layers” and “Duplicate Layer”.

This… does nothing to the photo. But if you look at the layers window, you can see that there is now a second entry.

Make sure that the upper-most layer is selected, because we are finally going to make that colourful photograph a lot less colourful.

To make the image black and white, go to the menu bar of the image (who by now is an old friend) and choose “Colors” and “Desaturate”.

This causes a new window to open asking you “How do you want to make this image black-and-white”.
Actually its asking “How do you want to make this layer…”, but you’ve probably going “I don’t know which one of Lightness, Luminosity or Average to pick”.

Well don’t worry about it. Remember how I said that selective colourisation works best with a strong colour. You will want the maximum contrast between your selected (eventual) colour and its surrounding monochrome. So at the moment you can’t know. If you want to be really sure, you can always take the same image make multiple layers… and try out the technique on them all. Then you can pick the best looking one. Or make use of “undo”.

Anyway, pick one.

You now have a black and white image. But if you look at the layers window you can see that the colour photo is still available. (You can click on the green arrows at the bottom of the layers window to change the view between the black and white image and the colour one).

So now its time to put some colour back. You don’t use some colour selecting tool, but a mask.

Go back to the picture’s menu bar, and choose “Layers” then “Mask” and “Add Layer Mask”.

This causes a window to pop-up. The default here is “White (Full Opacity)”, choose that one and click “OK”.

The way we get the colour back is to “cut” through the monochrome image. If you look in the layer window you can see that there is not a white rectangle next to the monochrome image. This is the layer. White here means that you can only see the selected layer. Black here means that the layer is transparent, which lets the image in the layer below show through. To do this we’ll use the paintbrust tool from the tool bar.

The paintbrush is the 4th from the left and the 4th one down. Pick this tool and you should see the selected (and alternate) colour for the brush. The topmost one should be black, and the one below it should be white. You can see the selected brush in the image above its called “circle 11″.

You can try out the process. Put you mouse over the image and holding down the primary button, scribble on the image.

See the colour show through?
You can easily get rid of this scribble by choosing “Edit” and “Undo” from (you guessed it) the menu bar of the image.

While the above brush is fine, you should really chose a fuzzy brush. Its more forgiving if you go over the edge of the item you want to return to colour. Click on the “image” of the brush and you’ll see a selection of brushes. Pick a fuzzy brush (you can switch over and over)

and go back to the photograph.
Before you start re-colouring the image, you had better zoom in (“View”, “Zoom” and either “Zoom In” or select a ratio) so that you can more accurately colour the image.

Now you can start to colour in your photo. Use small strokes, that way you can use “undo” as needed.

Personally I prefer to go along the edge of the part I’m colouring in. Filling in is the easy part.

At this point it might be worth looking at the mask itself. Choose “Layers”, “Mask” and “Show Layer Mask”.

And you can see the parts I’ve changed.

You can also see a little “scratch” next to the sleeve. That is a mistake for which I can’t use undo. I can however use the paintbrush to make that black mark white again.

Back on the toolbox, you can see a little two headed arrow where the chosen colours are. Clicking on this will switch the colours.

Now with my brush “painting” in white, I can erase the mistake.

Switching the colours back to black, I can cheat.
Since I “painted” the edge of parts I want to re-colour, I can use the “fill” tool (in the picture above it’s the left-most column, 4th one down with an icon of paint being poured from a tin) to, well, fill in a chunk of the image…

… and switch back to the brush to tidy up the image.

When you are doing the selective re-colourisation, I’d advise you show the mask so you can see if you missed a bit.

You can now turn hide the mask (“Layers”, “Mask” and select “Show Layer Mask” to deselect it) and see the image.

Using the green arrows in the layers window to see the full colour and selective colourised image to see if there is another part you wish to colour in.

Now it’s time to save your image. Go to the menu bar and choose “File” and “Save as”.
Why “save as”? Firstly, if you are using a borrowed machine, you need to make sure you are not overwriting the original image, and that the location you are saving on is going to be on your own removable media. The other reason? Well, GIMP uses the file name extension (the 3 letters after the main file name) to decide how its going to save the image. If you save with a “.xcf” then you are saving n GIMP’s own native format. And it retains all your layer information if (like me) you will need to return to the file to edit it. If you don’t want to return to editing, then the file name extension of the image file sill still be set to the file type you opened (in this case a JPEG or .jpg file).

When you save after editing you are going to get a few warning boxes.

This one reminds you that you are going to loose your layer information. Since that’s the point choose “confirm”.

This one tell you that since a jpeg can’t handle the transparency its going to turn your multi-layer image in to a single layer one. Again that’s the point; choose “export”.

And finally (you will get this one)

The scroll bar on the top wants to know about compression. Since these images are going to be for printing, choose 100%. However if you are just trying out the technique, (and the 100% reslts in a large file size) you can leave the quality levels at a lower setting. Once you are happy with the quality setting click on the “Save” button.

We now have a selective coloured image.

If you look at the pointing hand, I didn’t colour in the cuff. This if you remember is the difficult part.
I could use the same careful drawing technique that I used before (or just leave it as it looks OK) but I won’t.

I’m going to leave it to you. This is the reason for the half post. I want you to tell me the techniques you used to colour in this part. Leave a comment below or e-mail me I’ll post the replies next week.

Take care,
and do your homework,
Will Knott

If I have the timing right, this will be uploaded while I’m at an interview, so it seems timely.

My CV as a tag cloud…

Robin Blandord came up with the idea to run his CV through Wordle.net to see what it would look like as a tag cloud. It’s been repeated by Ellybabes, Sinéad Cochrane and Paul Browne. Sorry that it’s in simple black & white and not visually pleasing as the others; I thought about adding it to my physical CV or embed it in the files to see if there was a change in uptake. Given that most (if not all) recruitment agencies parse the résumés they recieve through something similar, it might make sense. Wordle is a Java application which parses text, ignores common words, and creates a tag cloud. The size of the word is proportional to the how often it occurs in the CV.

Putting something like a résumé through it might take a bit of tweaking. Plurals are (currently) counted separately, and other common phrases show up. I needed to tweak my CV to stop my address showing up in the tag cloud for instance.

The other thing is how “management” looms large over both “software” and “database” (with “sql” peeking in there). I didn’t expect that, nor expect to see something similar on the other technical people’s clouds.

Does it give an accurate picture which should be used as a reflection or consideration for employment, or is it really easy to “game” the results to create an artificial profile?

And given the full social media treatment, does a tag cloud of a persons delicious account cast a different reflection than their résumé?

So then,
You hiring?

take care,
Will Knott

Related articles by Zemanta

• What’s a resume, these days?
• Great Graphic Find: Wordle
• Twitscoop Rolls Out Real-Time Tag Cloud for Twitter
• Employers slow to realise the benefits of web 2.0

The Crabling Otter that is Darren confessed that he carries a man bag. Needless to say Granddad is disgusted (so no surprise there). However I have a bag with me now. Its a backpack (no single strap stylish bags for me) filled with bits and pieces of my pack rat existence, and its a school bag again. I also have a Hermès Kelly bag, but more about that at the end of the post.

Image via Wikipedia

The bag is filled with the usual important bits and pieces. USB Memory stick, the non tech part of a MMC memory card, cables, book, screwdriver, multi-tool and pliers, cables, headphone bits, cables, book, pen and paper and of course cables. And books. And bits and pieces for my class.

One of the guys in my class asked my why I didn’t blog about, well, the class (yes I’ve gotten new readers, and I’ll link to them when I get them blogging) and I couldn’t see why not. I’m studying Software Development in Fás. And yes, I’ve been programming for years, but given that my background was in mathematics, and most of my working life was making databases do things that they shouldn’t do, I’ve not studied computer languages properly. Yes I’ve inhaled manuals due to a particular project requiring it, or simply because I wanted to, but not formally. I think my pfd collection shows that.

And yes I will blog if I get any formal certification along the way. And no I’m not referring to a diploma from Miskatonic University (got to love Lovecraft), which in also in my PDF collection along with my e-text books and paper models.

Which brings me to Hermès and the Hermès website which thanks to a link on the Paper Forest site I found out is offering a downloadable set of paper model Hermès Kelly handbags. The real one is a very expensive hand bag named after Grace Kelly (hence the name). These paper bags are an official release from the Hermès web site and comes in eight versions including a plain blank one for customisation.

I wonder if I can write code on it?

take care against paper cuts,
Will Knott

As much as I love old games I just can’t love the flash based retro games brought to the web by Paul Neave (but I will admit to being impressed by his flash planetarium) . The games include Space Invaders, N-blox, Asteroids, Simon, Tic-Tac-Toe, Hexxgon, Frogger and Snake (as in the mobile phone game that Nokia phones had so as to spend many hours not talking on a bus). At the moment I’m smitten by Chain Factor.

Source: WikipediaChain Factor is an apparently simple flash based game. You are dropping numbered (and blank) discs on to a grid. If the number on the disc corresponds to the number of touching discs in that row or column, then all the discs of that number in the row and column vanish and the remaining discs drop down. The blank discs reveal their number when a 2 discs touching it has been removed from play. Levels progress with a new line of blank discs sprouting up on the bottom row. Seems simple, plays hard and is a touch too addictive. So be warned.

And while I’m typing about games, I want to bring up a recent discovery… Ben “Yahtzee” Croshaw of “Zero Punctuation” game reviews fame created “The Art of Theft” a 2D cat burgular stealth adventure which is an awful lot of fun, a little ZX Spectrum or Commodore 64 / Vic 20 ish fun. No wonder tape died.

take care,
Will

I like listening to thing I shouldn’t. Sometimes I get things for “overheard” posts (which I really ought to do more of) and sometimes I hear interesting things. Like Bernie Goldbach talking to his students about the work of Flight 404, who is also known as Robert Hodgin.

Image via Wikipedia

As well as the ubiquitous day-job, he also spends time playing/working with Processing. Processing is an open source, multi-platform programming language and environment for people who want to program images, animation, and interactions. It is used by students, artists, designers, researchers, and hobbyists for learning, prototyping, and production. It is also a visualization programming language. One for creating images. However it uses Java as its base, so it might be correctly described as a programming extension library (Processing code can be exported as Java) and IDE for Java.

I know, another programming language library. Ho-hum. However this one is, well, bloody beautiful. Take a look at what you can do with it (I’d advise using full screen).

Solar, with lyrics. from flight404 on Vimeo

The language is designed for work like this. I’d love to see it mixed with a physics engine. Robert goes through the details of how he created the above demo and it seems a little intensive, but fun. Given that unless things change I might be back at programming school soon, now is probably not the time to start a new language (or is it exactly the right time?). I would love to play with this visualization method on a couple of mashups. I’m thinking about music, but I can see it working on time related information such as relative population sizes or economic data.

What can I say, my mind moves in odd places sometimes.

take care,
Will

A long time ago, at MashpCamp I thougth about combining a Google Map with the geodata of pubs and a ratings system to generate a “Rate-my-pint” type application, which I’m sure a brewery could use for customer support and ensure that their product is being served properly. This is probably more useful.

Image via Wikipedia

I have no idea who made the Dublin Pub Crawl map, but I’m impressed. Essentially it’s a Google map with route calculations which allow you to create your own own pub crawl through Dublin City. However I’m not too sure how to get the data off the computer and in a format which would survive a “12 pubs of Christmas” tradition.

Choose an area, like Temple Bar or Dublin 15, pick some pubs you’d like to include, if you like. And you can edit the maps and add other pubs.

Like those outside of Dublin?

I suppose it’s all about how intelligent the routing code is, but would adding a bunch of Cork, Limerick and Galway pubs just mess things up, or make the map more useful for those who don’t want a pint of plain in the pale? And it it works, then we could have “centres of excellence”, in drinking.

take care,

Will

LOL cats have been doing the rounds… now a .NET compiler for the LOLcats language

Oh dear….

What would Rabbit do? Do Rabbits and cats get along?

Still, I have my spleen

Will