Internal Server Error

This blog has moved

2010-04-02T11:27:00.001+01:00

This blog is now located at http://htmlerblog.blogspot.com/. You will be automatically redirected in 30 seconds or you may click here. For feed subscribers, please update your feed subscriptions to http://htmlerblog.blogspot.com/feeds/posts/default.

So I had an idea.....

2010-01-14T20:24:00.001Z

A while ago I saw a BBC show with Monty Don travelling the globe looking at gardens around the world. He visited Cuba to look at the fantastic urban farms and community gardens there which are going a long way to helping feed the people of Havanna in a way they can afford.

Then I read a few things about the urban farmers and community gardeners in Detroit and Todmorton where wastland is being turned into gardens to feed the residents. I guess watching people like Hugh Fearnsley Wittingstall, epecially his landshare intitive and Carol Klien on tv and reading their books have helped too. Definately having our own garden helped immensley.

The problems with gardening are various. One, its hard work at least at the start and at peak times of the year too. Two, you either have too much stuff or not enough. Three, economies of scale come into it, bigger gardens are cheaper to run especially for things like compost etc. Four, you cant always be in your garden when you should be to water, plant or harvest. If you could share these problems it would be good. Trade time, resources and money for quality produce.

I got to thinking. What might work would be a sort of co-op. You sign up to the scheme and say you will donate x ammount of land, y ammount of time and z ammount of money. This is a sliding scale for all three. If you can donate enough of any of one out of three you get the others free. So people with a large enough ammount of land dont pay or spend time. If you have no land and put in enough time you would get the benefit with no other outlay. Simple cash donation get rewards too in the form of the produce as in buying fruit and veg from a shop.

The intial investments of time, land and cash get put into a pool. Landowners get first pick of what gets grown on their land but have to accept that stuff will be grown that they don't necessarily want themselves. Cash gets spent buying tools, seeds and other materials. The land gets coverted into efficient fruit and veg plots by people volunteering their time. Things like compost, tools, seeds, transport etc can be done from a central location.

These then get maintained by the volunteers and harvested etc. Each landowner can take what they like from their own plot but only enough for themselves and their family etc. All surplus produce gets first redistributed amongst other people in the co-op so the people who are time rich get their free produce and the people who invest money get the produce they paid for. Anything left after that is sold on the open market. Profits are returned to the scheme for new tools, seeds etc.

The system has to be kept small in scale. The key to it is how much time is required to invest to get a unit of produce. In effect a wage for the volunteers. It has to be set at a level that encourages people to invest time and not money unless it soon becomes easier to go to the supermarket to get produce than through the scheme. There would have to be limits on the ammount of land, time and cash one person could put into the scheme though this may vary locally.

I guess to get this off the ground you would need to speak to someone with far more knowledge of economics than me. Then get some advertizing done. Then get digging.

Towards a more unified way of developing sites

2009-12-12T15:15:00.002Z

The division between designers and developers has existed almost from the birth of the printed word. Technologies have been produced to make information more rich and interesting from more complex layouts, pictures, and now video, sound and all kinds of interaction.

There is always some form of either a technological development being exploited by designers or a technology being developed to meet a design based need. In most cases it is likely to be a combination of the two.

The problem lies that designers and developers do not always see eye to eye on many levels. There is often a level of resentment in the relationship between the two. Developers view designers as head in the clouds types with no idea of how things get really done. Designers view developers as stick in the mud types resistant to change. They both talk about teamwork, Agile methodologies and co-operation but often the creation of a site ends up as a basically two stage process. The designers create a plan of how they want the site to look and work and hand it over to the developers. Then they may double check things at the end of the project or iteration etc.

This is just a waterfall process by another name. The designers "hand over" their wireframes etc to the developers and say "get on with it". They might have used Agile to develop the wireframes and the developers might use Agile to make the actual site but there is still a massive discontinuity between the two parts of the work.

Often the designers and developers are in separate teams and entirely separated in different offices etc. Even in my current job the designers are at one end of the office and developers are in the other end. Often the designers can be in a separate company to the developers.

I personally believe having worked as both a designer and a developer in different roles that close communication between designers and developers is vital. Communication is all about the exchange of ideas and more communication means more ideas. Maybe not better ideas but there are still more ideas than there were.

Designers and developers should be involved at every stage of the entire process of creating a site from its conception right through to its on-going use by real people. Database developers need can help plan the basic functions of the site. It will make their job of developing a schema easier and they might well be able to say "if you want it to do XYZ to make things easier" and create something entirely new that the designers had no idea could happen. That is only one example. The opposite of course can happen as well. A designer can come up with a requirement that a developer would have never have thought of.

At the moment the process is one way. Designers design and developers develop. When either group crosses the boundaries into the others territory then conflict occurs. There needs to be a breaking down of barriers to some extent. Of course there have to be specialists. Its crazy to expect a graphic designer to be a DBA as well.

Processes like SCRUM and Agile help with this to some extent as they should get people working together. However this does not always work for the reasons outlined above.

I think that there are enough similarities between different roles to implement a sort of eXtreme Programing technique combining developers and designers. This is based in part on my experience several years ago. I was working for a small college as a web developer working with a designer to create all kinds of material based around HTML content.

We quite often worked on interface elements of a project at least as an XP team. I would be working in Dreamweaver, Photoshop or the like with the designer sat in the "co-pilots" chair and between us we came up with a solution for a problem. At the time it could be hell and led to some quite spectacular arguments and I am not advocating this exact approach but it is a good starting point for a discussion.

What I am proposing is a kind of buddy system. There is enough common ground between several types of jobs in both the design and development fields to make this work. For instance a information architect designing the high level specifications for the site could work alongside a database developer who models how the site is represented in a database. They are both concerned with how the site is organised and that gives them common ground.

The obvious team up wit the UX and front end developer working closely with each other. The developer can advise the designer on what is possible but the designer can come up with new ideas that the developer might have missed.

The problem lies with who to team back end developers with. They are not a natural fit with graphic designers. They may have to be involved with some aspects of UI development but they often avoid this sort of work. There seems to be more correlation between the sorts of structures and organizations that an information architect creates and object orientated design than anything else. Taxonomies, Labelling schemes and metadata all have their place in the object orientated world.

Some other functions a site may be expected to have may be very interdisciplinary. Search for example will involve aspects of information architecture, user experience design, front end development, back end development and database design. If any one of these is omitted the function is likely to be less useful to the end users.

A personal example is a good way to illustrate this. Recently I was asked to provide a series of wireframes and other UX documents to a recruitment company for a site that its candidates could use to look for positions, clients could use to look for candidates and its consultants can use to match candidates with vacancies. By co-incidence the first real commercial website I produced was for a different recruitment company nearly 10 years ago now. Back then I looked at the companies paper based forms and processes to create a database schema to store the information they used and built the site from the back to the front so to speak. This time round I evaluated the companies existing site and interviewed its staff and also some candidates and clients about the site to conduct a basic usability study and expose some new features that would be useful on the site. These data were then turned into personas, sitemaps, wireframes and other documents which were turned over to the clients developers to implement.

Essentially the same process took part this time round as 10 years ago however. The difference in the two is 10 years ago I was more of a developer and understood more about databases that information architecture and user experience design. Now I understand far more about the design side of creating a site than the back end development of a site, certainly I know more about usability than SQL now. What occurred in both cases was that I modeled the users needs and requirements. I just used two totally different methods to do so. However this time round I could apply the knowledge I gained 10 years ago when designing a database to model a recruitment company to my current design problem.

The key area of the site as with many others is search. Candidates want to find jobs, consultants want to find clients for candidates and vice versa. Clients want to see a good pool of candidates. All of this when you get down to the real nitty gritty is a well designed and executed database schema and set of queries to interrogate it. However it is down to the user experience designer and information architect to make the search as easy to use as possible. No one wants to type in SQL queries to get their data out. (apart from real DB geeks) Users demand power and flexibility. They want to choose from a small list of options and sprinkle a few keywords to find their dream job. It shouldn't take long and it shouldn't be difficult.

This is where good design comes in. Both the DBA and IA have to accurately model the users needs. They are both using different forms of design to achieve that end. The DBA has to work out a schema that is both accessible in terms of the ease in constructing SQL queries to interrogate the data within it and also usable in terms being performant in returning data quickly. The IA/UX designer has to model the users needs into in effect a web form that generates the SQL queries that the DBA uses to interrogate the database. This has to be accessible in terms of ease of use and performant in terms of enabling a user to get the results they requested.

Its not implausible that the IA/UX designer may model some user behaviour that the DBA had not thought of. Similarly the DBA may be able to spot and model some user behaviour that the IA missed. What is often overlooked is that when two experts work together on a common problem from different angles they are able to collaborate to gain insights that they would not otherwise have come up with independently. It might be possible to discover a new way of requesting data that might have been overlooked. A unique selling point perhaps or certainly a competitive advantage. If the "designers design then hand it over to the developers" model was followed in this case they this synergistic process would be missed out on.

Teamwork is the core to all of this. Maybe not the extremes of XP or lesser peaks of Agile but good communication and the sharing of ideas and concepts cannot be a bad thing. When people share they create something they would not be able to do individually.

This is both inspiring and though provoking

2009-11-14T11:55:00.003Z

I came across Leagh Buley's presentation on Being a UX Team of One the other day. I have to say I watched more or less every minute of the video and followed the slides too.

Leah seems to have experienced some of the same problems I have come across when working as the "Front End Developer / UX guy" and has some great ideas for overcoming them. So often someone has come up to me and said "can you come up with a page to do XYZ?" and I go away and think for a bit and come up with one design solution. In the bad old days I would have started coding straight away. Now I can write html and css quickly but not that quickly.

Then I got into using graphics packages to create high fidelity mock ups before I started coding. Almost the old slice-n-dice photoshop table layout days all over again. Of course I obsessed over details and didn't come up with a coherent design.

I was never that keen on sketching UI with a pen and paper. Mainly because I am really bad at drawing I think. It brought back bad memories of sitting in an art class aged about 13 with no idea what I was doing there. Then I joined I ♥ wireframes on flickr. Yes there are some very beautiful wireframes created in graphics packages but the most interesting ones (IMHO) are sketched.

Leah definitely is a big fan of sketching and I think am now converted. The main thing I got out of Leah's section on sketching was that you "brainstorm, a lot". Produce many ideas and mix and match. Let them evolve. Don't limit yourself. I think I might have frustrated my co-workers quite a bit in the past by coming up with one design and "evolving" it without really asking or telling them. Evolving the design on paper stops this sort of problem.

As an aside from all this there must be something wrong with computer based UI design in general if it limits this sort of creative process or other similar creative activities.

The next stage of working as a team of one is a bit contradictory. Basically don't be a team of one. Assemble an ad hoc team around you. Basically get some help from your co-workers. This isn't really a team of one thing if you ask me but there you go. I can see the point that your co-workers will put their oar in anyway so why not get their input. From personal experience I think this can go one of two ways. Either they will tell you to sod off and stop asking you to do your job for you. Or they will jump in with tons of unhelpful suggestions that stop you working that way.

I guess the trick is to fall between these two extremes and get coders to help you constructively. Leah and Adaptive Path have a range of techniques like sketchboards, open design sessions and the like. One of my favourite ideas is to decorate your workspace with your work in progress to get people looking at your work and commenting on it - hopefully constructively. The idea of passing someone who is commenting on your work a pen and saying. "Draw your idea then." is great.

The final section is a bit ass about face I think. I would be coming up with design principles at the start not the end. If I had to for instance come up with a set of designs for a hotel's site I would try and figure out the design principles at the start and then use them as a filter for ideas right from the start rather than at the end. Though that could mean rejecting ideas that don't fit the preconceived ideas which is a bad thing. So maybe this isn't a bad idea after all.

Watching the video and googling around the subject has really made me think. I can definitely improve my process for coming up with a design from this. I might not agree with all of it but hey that is bound to happen.

UX isnt just for the big boys

2009-10-24T16:41:00.002+01:00

Recently I have been looking at a lot of relatively small scale web development companies. They produce a whole bunch of sites for small and medium scale businesses. They all seem to view UX as a either part of the graphic design of a site or an unnecessary luxury that they can not afford.

I would personally argue it is something that they cannot afford to be without. It adds an extra selling point to their pitches at clients. It might require explaining a few extra things at the same time but at least it should give their potential clients something to latch on to compared to all the other competing agencies. Its not a unique selling point but at least its a different selling point.

Small agencies seem obsessed with graphic design and SEO. Its almost as if nothing else matters to them or their clients. I would argue that they often fail to meet their clients expectations because they don't consider the user experience. Getting things like e-commerce right is not a matter of fitting things into a CMS/e-commerce system that comes off the shelf. Its a complicated process that needs tailoring to individual sites customers needs. Its in areas like where UX comes and has a huge potential to make a difference to the clients bottom line and the agencies chances of getting more business.

If a small 10 or 12 person agency had one UX person working for them I would personally be prepared to bet that they can pull ahead of their rivals as it would give them a competitive edge. UX is not for the big boys. It gives the little people the chance to compete by giving them more tools to work with.

I still use my grandfathers tools

2009-08-31T16:00:00.001+01:00

I still use my grandfathers tools
Originally uploaded by Rob 'n' Rae.

This picture means so much to me. Every time I use one of these tools I get a sense of the past and also responsibility to keep them in good condition for many more years of use.

on line identities

2009-07-06T10:38:00.002+01:00

I just realised its very easy to find me on line. I nearly always use "rob_cornelius" as my username, unless its already taken or the site I am signing up with wont allow underscores in user names. So I am rob_cornelius on hotmail rob.cornelius on google mail, rob_cornelius on del.icio.us, robCornelius on digg, rob_cornelius on steam games even. etc. etc. etc.

Now hardly anyone uses their real name on line, especially for things like games where everyone else is called Psk0killa or something similar. I think sub-consciously I must have decided that its hard enough to find anyone on line if you are not searching facebook or something like that. I don't mind admitting I am who I am so I use my own name.

Zero configuration wireless networking the easy way

2009-05-28T21:17:00.002+01:00

For some reason since the staffroom was re-decorated the wireless signal in the study has become too faint for the little wireless dongle to pick up. So.... I bought an extender relay sort of thing. It was a nightmare to set up. Basically if you read the box you plug it in and press the auto-configure button on the side and that's it. Rubbish. Don't bother with the set up disk either as that is even worse. It crashes regularly and doesn't work at all. Get on the net and find the real instructions.

Then if you put it in the living room where it can get a good signal the main pc in the study still wont get a good enough signal. So... put the wireless extender in the study. Now it cuts in an out with its link to the main access point.

So now I got a more powerful wireless dongle with a proper aerial for the main pc. That let me move the extender back to the living room for a better signal. Even better I put it on the far side of the living room so I can now get a wireless signal in bed. Working from bed.... maybe ;)

Some things look easy on paper

2009-05-14T20:10:00.002+01:00

I have spent the best part of a week working on a drag and drop interaction for the site I am working on. It was one of those things that look relatively easy but turned out to be a nightmare.

The set up was in theory simple. Two <ul>'s containing <li>'s you can drag and drop around both within the lists and from one list to another. Not that hard you might think. The added spice was that the UX guys wanted a swapping sort of thing as you drag the <li> around. As you move over another <li> you see a sort of swapping animation. Not just on drop on dragging over every element.

That complicates matters somewhat. Especially when you drag from one list to another and drop. Does that mean you are adding to the second list or swapping the two around? Well it works that if you drag and drop on top of an element in the second list you swap, if you drop it between elements or at the bottom of the list you add to it.

Then what happens when you completely empty a list? That means invisible placeholders being created and destroyed as you need them. All kinds of complicated stuff.

Lots of work in the end to add about 30 lines of code to Evans original code (which didnt have to worry about empty lists and stuff). The thing about once a year in my ten year career as a web developer I get asked to do drag and drop interactions (thank you macromedia/adobe for saying its easy to do with Dreamweaver and Flash) and none of them work that well.

Drag and drop for web based applications always seems to me to be one of those bells and whistles that managers/ux people like but actually baffles many users. Its also a bitch to get working properly for the developer so I never think its worth the effort.

the wierdest IE7 bug I have ever seen

2009-05-06T21:19:00.002+01:00

Sadly I dont have a demo of this available right now... will see what I can do.

I had a simple task in work today a whole bunch of checkboxes and their labels on the screen. Clicking on some of the checkboxes causes others to be disabled thanks to some nifty javascript. Nice and simple.

However I was asked to find a way to make the disabled elements look "more disabled" as it can be hard to spot the greyed checkbox on the dark background of the site.

As the checkbox was immediately followed by its label I quickly came up with:

input[disabled=disabled] + label { color: grey }

To quickly change the labels of disabled boxes only using CSS. Works great in Firefox. Clicked on a checkbox in IE7 (dont have to worry about IE6 - Hurray) and nothing happens... Then I move the mouse and magically the labels change colour!

I spent quite some time playing with it and that is definitely what is happening.You can see the checkboxes become disabled but the labels don't change colour until you move the mouse. I have googled for ages and can't find anything on it at all. Might try a static test page rather than the rather complex html js and css that there is there now.

I guess that I could fake a mouse event to fire off after the checkboxes get disabled but its easier to change the labels style at the same time as the disabling in the first place.

Why for the love of all things nice and fluffy why?

its gardening time again

2009-04-29T20:20:00.002+01:00

If you take a look at our flickr photostream you can see we are back into the swing of gardening again.

Last year was a bit hectic with moving house, bad weather, cricket and lots of visitors so we did really get as much out of the garden as we could have done. This year, weather permitting, we should be able to enjoy it and enjoy the fruits of our labours.

So far in the veggie patch there are:

Salad leaves of various sorts
Rocket
Chard
Perpetual Spinach
Spring Onions
Sweetcorn
Spuds
Squash
Carrots
Radish
Peas
Garlic
Beans
Courgettes

The soft fruit list isnt bad:

Strawberries
Raspberries
Blueberries
Blackcurrants

In the greenhouses there are:

Tomatoes
Sweet Peppers
Chilli Peppers
Cucumbers
Ginger
Lemon Grass

Then for herbs we have:

Thyme, 3 or 4 sorts
Coriander
Basil
Mint
Oregano
Marjoram
Chives
Sage

So all in all not a bad crop if it all works out.

The flower garden is coming on great guns too. We have planted about 1800 bulbs or so over the last 2 years and a good percentage of them seem to be ok. The tulips are looking great and the aliums will be out soon too. Hopefully this year some of the perennials we grew from seed will be doing better. Hollyhocks and stuff like that. Also this year we have decided to have a real go at the "patio" area with hanging baskets and containers all over the place. Watch this space and flickr for more.

The best thing about the gardening is its totally relaxing. Even when I am working hard digging or something like that I feel totally relaxed. Its also having a sense of that its ours. Living in the school takes that away from us to some extent. We know the flat isn't really ours no matter what we do to it but no one wanted our garden before we started and no one thought it was possible to do anything with it. We have turned it from a patch of head high weeds to a "little oasis" as Sarah was saying the other day. There is no finer pleasure than looking at something beautiful and saying "I made that" apart perhaps from eating a great meal that you have cooked yourself made with ingredients you have grown yourself.

things have moved on since I was a lad

2009-04-25T15:24:00.002+01:00

I just downloaded some pretty fractal generator for my nice big quad core maxed out box. Its blindingly quick. Even full screen at maximum resolution it takes about 1/4 of a second to draw the entire screen.

When me and my brother had an amiga back in the late 80s we used to wait hours for it to render an 800x600 mandlebrot set or even leave it running overnight.

Of course thats nothing. I have worked with several people who started on paper tape or punched cards. And to top it all one guy who used to come in and help out with the computers when I was a field ecologist for the Somerset Wildlife Trust used to actually work at Bletchly Park.

How to filter out a lot of the crud that pollutes google search

2009-03-27T13:09:00.002Z

Don't search on google itself. Search on something like del.icio.us instead. That way you are searching only sites that other people have found interesting or important enough to bookmark.

Hopefully this should filter out a lot of the useless adwords paid results, "price comparison" sites, scam sites, lolcats, porn and other annoyances for you and generate you more relevant search results automatically.

This should be a huge resource for yahoo, who own del.icio.us but I guess they want to get their advertising revenue so they are not going to push it.

designing for localization

2009-03-25T21:31:00.002Z

I am working on a large site currently. One of the design requirements is that the site is easily localized as versions of it will be in use in various countries.

JSP/Java is very good at doing that sort of thing but its very important to plan carefully. Every single piece of text needs to be configurable. J2EE does this by using resource bundles. Everything is a key/value pair in a .properties file and <fmt:message> tags insert the text into the jsp. Then when a user comes to the site and their browser is set to a different locale the J2EE wizardry grabs the text from the right .properties file and the user sees the language they want.

That's the theory. And to be fair it works quite well. The trick is to plan in the localization from the start. You can't bolt it on at the end. You cant use an <input type="image"> tag to have a pretty looking button as that would mean the text would be in an image file and not localizable. So instead wrap a <button> tag with a <div> tag. The <button> tag displays the text in a nice localized way and the <div> has a pretty background image with no text.

Of course it can all come crashing down around your ears. For reasons best known to themselves the German language wont use one syllable when half a dozen will do. So your button will have the right text but it will overflow the background image. There is no real way around this though. The only thing to do is test, test and test again.

doing my bit for the internet

2009-03-07T09:50:00.002Z

I set up a flickr group a while ago for the staff at the school where I live to share pics of the school. We don't post pictures of the children for fairly obvious reasons. I set the group up so you had to ask to join just in case.

Then the other day I had an email saying some random person on flickr wanted to join the group. I checked out their profile and they had no pics at all. I checked out their groups and they had obviously searched for every group on flickr with "school" in the name and joined all the groups they could.

Very, very odd indeed. I mentioned it to Rae and we decided this was worth doing something about. I reported the person to flickr's abuse system and they "are no longer active on flickr". Good.

Oh I have yet another new job. I am now contractor scum working for Qualcomm!

Its been a very busy few months

2008-08-26T13:05:00.003+01:00

Since Easter I have changed jobs twice and moved house complete with redecorating the new place from top to bottom. Not to forget looking after Joshua and Oliver for quite a while while Julian and Heather packed ready to go to Canada. Some of the changes are detailed in a Previous post

Lets just say the job with Patersons didn't work out and was very frustrating. Then one day Rae was on MSN and said she had seen a web development job in the local paper. She said it was for a company called Lightcurve and the contact name was a guy called Stuart Lee. Well I went to school with Stuart and after an interview and a phone interview with one of their partner companies in Dublin I got the job.

Giving my notice in to Patersons was definitely a good day. Even if I had a raging hangover from drinking the best part of a bottle of champagne to celebrate getting the job. It probably turned out for the best as they were laying off people as I left.

I finished with Patersons on the 31st of July on a Thursday. Had the Friday off and met up with Neil, Sarah and the twins then and then started work on the Monday.

The new has been pretty cool. 20 minutes to drive to work is nice. A converted barn with air con instead of a tin shed to work in is nice. Not drowning in paperwork is nice. Not having 4 hour meetings at least once a week to report on "progress" is nice. Having a decent pc with two monitors instead of a laptop to work on is nice. Being able to get up at 8am, have a shower and still get into work with plenty of time is even better.

The decorating wasn't too bad. To be honest Rae did most of the painting. I painted a few ceilings and undercoated one room and she did the rest. Best thing in the world ever was the roller on the telescopic pole. It makes thing a lot easier. We spent yesterday putting up shelves and stuff. Tonight its pictures. Rae is busy making curtains too. We still don't have the lightwell sorted out but that can wait.

It was great to have the boys over too. Again Rae did most of the hard work as I was working. They can be a handful so again she did really well. I know Oliver will love his toolbelt we bought him for his birthday. He will be demolishing the furniture in Canada with any luck.

Lets hope things settle down for a while at least.

taking it slow

2008-07-28T12:49:00.002+01:00

We had a very slow weekend. Though come to think on it there was quite a lot of driving around and even the odd bit of shopping. Perhaps it just felt slow due to being incredibly relaxed for a lot of it.

Saturday was a quick trip to Yeovil in the morning then off to Warminster for Rae's Uncle David and Aunt Wendys 25 wedding anniversary . Lots of sitting around in the sun drinking (non-alcoholic for me of course) and chattting. Then over to Kellys for a barbecue the highlight of which was seeing Kelly turn her grandfathers head into a large ice gem with some spray cream. Oh and 2 1/2 pints of Pimms was very nice too...

Late start on the Sunday too then off to Dorchester to pick up a bookcase Rae bought on ebay for the new flat. Back home via Tescos and B&Q to get some odd bits and bobs. Then the rest of the day was spent floating in the pool, relaxing in the garden and watching a hot air balloon launch from one of the school fields. Oh and cooking some very nice ribs on the barbecue. We stayed out in the garden until it got dark and it was great. Just read the paper, chatted and relaxed.

Next weekend we have Neil, Sarah and the twins visiting on Friday then Julian, Heather, Joshua and Oliver arriving on Saturday for a few days before they go back to Canada. I don't think we will have much time to sit in the garden and relax then.

sometimes you cant see the wood for the trees

2008-06-03T11:57:00.002+01:00

The cyberpunks in Toxia have been looking at setting up a centralized database of information for some time. As I have a few skills in this area I was looking into it for them.

For some reason I couldn't think straight and was playing with all kinds of strange and exotic toys like RDF triplestore databases and the like. Then last night it hit me. A wiki is perfect.

What really annoys me is that I have used wikis extensively in the past both professionally and personally. Why the hell didn't I realise that the solution was so simple?

Its been a while

2008-05-28T19:57:00.003+01:00

But I am back.

One of the main reasons I have been away is that I have a new job. No more being the webmonkey in the corner at ingenta. I am now the slightly more senior webmonkey in the middle of the room at Patersons.

Ingenta made the very generous offer of "move to Oxford to work, no extra pay and no chance of working from home or take redundancy". Let me see now.... I could barely afford to work for ingenta despite having hardly any bills when I was in Bath. So... "I will take the redundancy please".

So after a bit of gardening leave and a few very annoying interviews and experiences with recruitment agencies I am now working for Patersons developing material for their websites and internal tools.

It's been interesting so far with a lot of new systems to learn. Hopefully some of the experience I gained at ingenta will be useful too.

Of course the other thing to talk about is the garden. The gardening leave did come in handy and its looking really good. Had some great rocket out of it the other day as a first crop from the veggies and some of the other stuff wont be far behind. The flowers were great this spring too. Well last year it was basically bare earth in the spring but this year we had tulips, daffodils, crocus, anemones and aliums in profusion. The tuilps were especially showy but I really like the aliums they almost look like fireworks bursting. Hopefully we will have some more food to eat soon. I reckon either pea shoots, radishes, lettuce or maybe swiss chard will be ready in a week or so. Possibly even spring onions.

We have been using our ingenuity too. I bought a little drip feed irrigation system from Tescos for a tenner and use that to water most of the containers. Then we noticed a length of hosepipe that had been lying around for ages. No one wanted it so I stabbed lots of holes in it and its now laid out on the veggie bed as a soaker hose. I can do most of the watering whilst sat on the new bench or better yet laying in the hammock I got Rae for her birthday.

Freecycle really came up trumps too. I got about 1/2 a ton of gravel for free to go along the bottom of the garden so we dont get muddy going to the new greenhouse and the wormery. Then Rae picked up a 3 tier planter thing and some hanging baskets the other day.

The other big bit of news is that we are moving house. Without leaving the building though. Hopefully in the middle of August we will be moving into another flat in the school. Its bigger and also more private. No more living next door to a bunch of screaming schoolgirls. Rae already has plans for decorating and has sent off for about 100 colour samples, or so it seems. The interesting thing is it actually has a lightwell in the middle. Nick and Karen dont use it for much but we are going to use it for houseplants and bringing on early plants for the garden.

Well thats quite enough of all that for now.

When to use programmatic constructs to get around browser failings.

2008-02-19T18:38:00.003Z

Often when I am developing a page there are things I could add to styling of the page to increase its accessibility or graphic design. Simple things such as highlighting the first paragraph of a section by changing its color or font weight. These can add immensely to the visual impression of a page with little additional work.

In theory that is. CSS provides a simple pseudo-class for the above example :first-child. IE6 doesnt implement it. Now I am left with two realistic alternatives. Either don't include that styling element and reduce the functionality of the design or work out a way to programmatically replicate that functionality in a way that IE6 can understand.

In the :first-child example its might be possible to achieve this if for instance the text for each paragraph is being individually pulled out of a database and looped through in a templating system to create the mark up. All that is necessary is to check if it is the first iteration of the loop and add a class attribute to the <p> tag which will hook into CSS to add the styling.

This is not always possible though. Perhaps the entire text for the page is stored en masse in your CMS system and its inserted into the mark up monolithically. You now cannot add in that class to find the first paragraph using your templating language.

Ok then, now what. Javascript to the rescue. You can run a bit of javascript as the page loads which selects the <div> tag representing the section, loop through its children, test to see if each child element in the DOM is a <p> tag and if its the first child. Thats messy, complex to write and error prone.

So, more javascript to the rescue. If you use a Javascript library such as jQuery which uses CSS selectors to attach actions or functions to elements in the DOM it becomes easy.

$('document').ready(function(){
 $('div.section p:first-child').addClass('emphasis');
});

does all the hard work for you.

Of course jQuery code runs perfectly well in IE6 so now you can use CSS selectors which never implemented them in the first place. There is the overhead of loading jQuery (or the library of your choice) but that is offset by caching and the like.

jQuery and its cousins do not just enable developers to create more interactive and engaging user interfaces they also enable designers to create more attractive and therefore engaging designs that will work in a wider range of browsers than ever before.

got my shiny new laptop

2008-02-11T19:49:00.000Z

After a bit of negotiation with the insurers my nice shiny new Acer 5520 arrived today. Its only the base model so no kill em all graphics card but its looking good so far.

The main thing that you notice apart from the wierd curved keyboard is how quiet it is. My old laptop was forever spinning up fans and stuff even under minimal load in windows. This one barely makes a noise no matter what I do to it.

Its also a little smaller and a lot lighter than the old one. The power supply is tiny too.

I have spent the evening installing stuff while waiting for the windows updates to come down. One thing thats coming off asap is Norton Antivirus which is as bad as having a virus in the first place. All I need now is CS3 and the stuff to connect to work and I am done I think.

Today is my lucky day

2008-02-07T13:05:00.000Z

I bought a kit kat today that had no wafer in it. Chocolate all the way through! Yum!

I found a use for AJAX on a site!

2008-01-29T12:04:00.000Z

I was ordering some seeds for the garden last night. Both of the suppliers websites I was on had a very handy feature. They had a page which assumed that you had a copy of the print catalog in front of you and therefore had the various product codes etc. On the page was a simple form with around 12 input boxes for entering product codes and the quantity for each one.

This was incredibly speedy, no searching through long listings of varieties of carrots etc to click on the right link. A really good example of integrating traditional paper based systems and e-commerce.

But the one thing it lacked was any immediate feedback on the product code you typed in for each input box. You had to wait until you pressed "add to cart" and check the contents of your cart very carefully to see if you had made a typo entering in your product codes.

With a little bit of ajax you could check the product code when the box looses focus and return a little bit of text or even a picture to show what that code relates to. Less errors, more feedback and better interaction.

Starting to like javascript again

2008-01-16T19:10:00.000Z

It used to be "javascript is evil, its only used to make bells and whistles that dont do anything useful".

I would jump through all kinds of hoops to avoid using javascript where it would have been very useful. I never really liked javascript coding anyway. It was always so repetitive with document.getElementById all over the place and lots of browser sniffing to get things right.

Then AJAX started making things even more complicated... and to be honest I don't see the point of most AJAX based sites. A lot of them use AJAX for the sake of it just like the javascript bells and whistles of old.

Early AJAX code was a complete mess too. Even more browser sniffing to be done and lots of complicated stuff to consider and code for. Along came javascript libraries that promise to take care of a lot of this stuff. I still didn't like a lot of them as they tried to force their own syntax which didn't always make a lot of sense. Some of them are like prototype / scriptalicious are tied to other frameworks etc fairly tightly. Others concentrate on silly interface widgets I never see anyone seriously use.

Now I have discovered JQuery and really like it. I like how you use css selectors to choose what to work with... I love how you can chain things together and I love how its possible to do AJAX and AHAH code with a single function. The onReady functionality instead of onLoad is fantasitic

I had to write a load of javascript today and it was a breeze. Select the element you want with CSS, attach a load of actions to it and impress some one in under 5 minutes.

the new cat is setting in nicely

2007-11-19T12:16:00.000Z

We acquired a new kitten last week. Her name is Millie though personally I think she should be called "special" as in "my mummy says I am special". Basically for it to be any more dumb it would have to be a rock. It walks into things that are there and spends ages apparently looking at stuff no one else can see.

On the plus side of things it has found loads of little things we lost around the flat. Also it produced a tennis ball from somewhere. No idea where though.

This post fulfills a prophecy of mine some time ago that all I would be able to write about would be kittens in this blog.