Archive for the ‘general’ Category

Another year gone by

Thursday, April 10th, 2008

A few days ago this blog celebrated its second birthday. It’s been a good year, with links to my article published by such notable authors as Luke Wroblewski and Ethan Marcotte – the same article was translated into Russian. The blog was also included in Planet Microformats.

I’ve written only 61 posts (down from 100 in the first year), gotten 134 comments (down from 155 in the first year) and more than 70.000 spam comments (up from 17.000). I have 56 subscribers according to FeedBurner, which is more than a 100% increase from 25 a year before.

Firefox is the dominant browser with 47%, Internet Explorer lost some market share and is down to 44% with Safari rising to 4% and Opera to 2.5%. Almost 90% of the visitors have Flash 9 installed, an additional 4.5% are stuck at Flash 8. Screens grew a bit with only 27% having 1024×768 or less (32% year before) and less than 2% 800×600 or less. Windows have a share of 85%, Mac OS 10%, Linux 4%.

Top content is still This page contains both secure and nonsecure items, Messing up the interface coming in at a distant second with merely a third of the visits.

Redesign

I’ve had a redesign planned since day 1, but as usual it took quite some time to get here. Since I’m not a design I couldn’t just create a fancy look with everything else left the same. The idea was to shake everything up and try to come out with a layout that would be worth redoing everything. When I posted about what a TV network / news page should have when being redone I was also setting my own targets.

Tableless layout

This goes without saying. All the pages should validate, although there might be some crap left.

Width

I increased the width to 850px with the addition of 140px to the left used only for design purpose, not content. The main content is only 410px wide which means you can easily read this blog on devices like the Nokia n800.

YSlow

Not really that relevant since the page only creates about 15 requests which might even decrease as I compact the JavaScript. There are a bunch of other requests made that are content related – images, Zemanta pixies, favicons…

Microformats

Every page is supposed to be valid hAtom and the About section is an hCard with adr and geo. Links to others are of course XFNed. The about page is planned to be an hResume.

There’s no need to talk about blog compatibility and screensaving for banners. I don’t use OpenID there are no log-ins.

Context sensitive

Markos recently pointed out the fact that not many pages on the internet make use of the possibility to change the page for different users and different context. This, of course, is not an easy job, especially if you’re doing heavy processing – doing this for each user might be a bit to heavy. It is something I think differentiates the internet from other media and should be used to provide a better user experience.

User based

Returning visitors of this page will get special treatment. Since they’re coming back they don’t need to see the About section and they will also not have banners displayed.

Referrer based

People coming from search engines will see their query parsed into words that will be colored for easier findability. They will be able to switch them on and off and be able to repeat the same query on the internal search with the last option available only if the referrer was an external search engine.

Location based

Even though many blog themes have the same sidebar for all the pages (as did the previous one I used) this doesn’t follow the normal architecture rules. The sidebar is normally context specific – in my case a regular user reading a blog post will currently see nothing in it (while others will see the About and the Ads.

Other candy

Grid and baseline

The page is set to a 7 unit grid, with units 60px wide with a 10px gutter in between. It’s also set to a baseline height of 18px that is respected throughout the page. Images and other non-text blocks on the page are corrected with JavaScript to a half of the unit (9px).

Links

The links bar will try to retrieve the favicon of the link – if it succeeds it will use it instead of the default icon.

Maps

If you click on the Google Maps link in the About section they will open as small inline maps with a link to open them in a new window.

Yet to be done

You might have noticed that post categories are nowhere to be found. I didn’t forget them I just found out that I’ve changed my categorizing pattern and had to rethink the display. I also wanted to add the Elsewhere section with links to or even content from other pages where pieces of me reside (Twitter, Flickr, Marela, Pownce, Facebook, LinkedIn, del.icio.us,…). Another thing missing is the Projects and Experiments sections, I’ll tackle these in a separate post sometime in the future…

Zemified

Announcing TechCrunch Slovenia!

Tuesday, April 1st, 2008

After a month of hard negotiation and work I’m proud to present:
TechCrunch Slovenia

The site is in Slovenian and you can read more at www.techcrunch.si.

Dear internet friends

Tuesday, December 25th, 2007

All I want for Christmas is to be able to buy a Nokia N800 Navigation kit. I’m located in Slovenia and the local dealer doesn’t have it, I can’t order it from Amazon (they don’t deliver electronics to Europe) and I can’t find it on Expansys (they were my best bet).

Be nice and help me!

Please?

Local web-apps

Monday, June 25th, 2007

Today a fellow web developer released a local web development job board called Unordered List. It was fun to be included in the testing while it was still in development. I made a bunch of comments and a lot of them got incorporated in the current release. It’s still a bit too monochromatic for my taste but still. I guess the development will not stop as at least a few new functionalities will be needed when the database grows beyond the front page – search and filters.

I’ve created a Yahoo pipe to get all the jobs in a JSON feed. I’ll be adding filters to the pipe as soon as company and the location of the job are available in the feed. This will make it easier to add a list of available jobs on blogs and other pages.

If you check Unordered List you’ll also see that I’m looking for an assistant. I’ll talk more about this in the next few days, I’ll just say that I’m after web enthusiasts who don’t yet know much and want to learn everything.

On a similar note, a local social bookmarking tool obviously launched recently. I have to say I noticed it through a Toboads ad network ad on my blog. I had to dig through the history to find it again but here it is – Zlitt. I have no idea who the author is though. The domain was bought through WhoisGuard and the app itself hosted on Dreamhost. A good way to hide your identity but probably not the best idea when you’re creating a local web-app. Who knows, maybe they’re going global…

In the mean time I put together a small one-page web-app to check your Lotto winnings – simply tell it what the winning combination was, the bonus number(s) and your combinations. It will show you what numbers you got right, but to know if you won anything you need to also tell it what winning means. I have it set up to the Lottery of Slovenia Loto, but you can change it easily. It’s not pretty, because I used the interface of my previous one-pagers – The pingerator and MonsterID.

I’m hoping for a lot of other web-apps from my fellow local web developers in the next months. Such that build on available APIs or are so specific to the local environment / community that they have a significant advantage over what others are doing. The local web-app landscape has been growing and maybe we’re in need of a page that will list all these web-apps in a way that is better than a simple search engine output. There’s an idea for another Web 2.0 page…

Who’s right?

Monday, April 2nd, 2007

Let’s say you’re looking for a piece of information and you only have two sources that offer it, which one would you belive?

  1. Wikipedia
  2. IMDB

Votes please. You can also add the correct birth date of Joan Jett.

The avatar fun

Sunday, January 21st, 2007

I’ve had a Gravatar account for some time now but there’s a problem with it – I don’t always want to leave the same email on all blogs. There’s another problem with the service – it’s offline currently, not accepting new users. Another problem and probably one of the reasons the service went down is also the centralized hosting of images.

For the reasons above I think Pavatar is a better solution. The idea is to retrieve the avatar image from the URL, not the email. You have a few options to provide a Pavatar link (http header, link in the html or simply /pavatar.png). The image is always downloaded from your site which might be a partybreaker if you’re a heavy commenter but better since you control the image at all time.

Today I stumbled upon MonsterID as a fallback for Gravatars. I loved it. I had to do something with it. I downloaded the script written by Andreas Gohr, released under the Creative Commons Attribution 2.5 License and changed it a bit. Well actually a lot. I added local caching and the possibility of adding other templates – I work at an interactive agency so I think I’ll be able to find a few characterbuilding image sets lying around.

The MonsterID service is here for anybody to use. It’s probably going to be as popular as the Pingerator.