For the impatient: Delicious Tag Manager
Having realized that I have way too much tags – something around 250 and around 1/3 of them with a count of 2 or less, I wanted to get rid of them. Unfortunately Delicious only offers a very rudimentary interface for this which needs a whole lot of clicks and confirmations and the only alternative I could find (Delicious Tag Cleaner is not working anymore).
I could remember that I once started working on a Delicious API library for Ruby (rulicious, originally hosted on Google Code). So, re-activating my old source code, moving it to GitHub I have a working concept for a tag manager (and the first state of a Ruby library for Delicious).
Unfortunately it doesn’t yet support OAuth for the new Delicious accounts, which are using Yahoo’s login. As soon as Yahoo’s API key generation is working again, I will fix this.
Where are you looking when you want to find a great restaurant or bar or just a take-away next where you currently are? Or no, let me re-formulate: What are your criteria for such a website? What is currently available on the web is for me a bit too cluttered, all the tagging, favoring, want-to-go, comments, etc. (besides some icons without any alt text, so how can should I know what 85 hearts mean?)
What I’m looking for (or willing to build) is a website with a very simple interface: a map, showing you current location (no problem with todays technology), spots nearby along with a scoring and the possibility to vote for a spot, meaning “I was there and I liked it”, kind of like digging the spot (although I shouldn’t use digged, is that copyrighted?) If you didn’t like the spot, didn’t feel comfortable or the food was not up to your expectations, just don’t vote.
There should be also the possibility to define a timespan for valid votes (or scores), so that you can judge the quality of a restaurant over time.
A little-known feature of the WMS standard is to add arbitrary parameters to the WMS query string. Not widely used, but with some great impact, you can add a time component to the WMS query and see data (tiles) for only a single point in time or a time range. And of course you can also animate it if the relevant infrastructure is existing. I tested my code on the US NEXRAD sources, using TileCache both for bandwidth and latency issues, and the performance was okay. As far as I know, TileCache doesn’t hold the pre-cached tiles in memory but reads them from the disk, which at least gets rid of net latency but could still be improved with keeping the tiles in memory. If you are interested in some code, you can find my ticket on OpenLayer’s Trac.
Recently I wanted to buy some new headphones. Not quite expensive ones, but also not so cheap (€100+). Just right for a guy with standard audio equipment.
So I went to our local electronics shop, which sucks regarding customer service, and – I wasn’t disappointed. After finding a responsible sales guy, this already took me two tries, he was okay-friendly and looked up the status of the headphones I wanted. Out of stock, for sure. Delivery data? Nobody knows. Awesome.
Next try: Austrian online shop of the retailer. MySQL error – 1045 : Access denied for user ‘atheas’@'localhost’ (using password: NO) (it’s still there, btw).
Ultimate try: Manufacturer’s website. Found my headphones, went to their online shop (which is on a different domain in a different web application, so I had to search for my headphones again, but that’s okay), put the headphones I wanted in the basket and clicked to checkout. Filling out my address data, I noticed that Austria wasn’t in their list of possible countries. Clearly an error, because they say they ship to the whole EU. Contacted their support stating my problem and got a reply within five minutes. Awesome. They fixed their website within some minutes, I proceeded my checkout process and was quite happy. Then I got the next email of their sales person, saying that the headphones I’ve ordered were out of stock and wouldn’t be available before end of March. Really seams that nobody wants my money.
I have to say that the sales person was really friendly and helpful – but still: How can you keep a product in your online shop and make active advertisement for it if it will not be available within the next three months? And regarding the local electronics shop: Why need I even pay for some guys who have no clue, are completely unfriendly and unhelpful?
In the last few months of the year, I was able to convince one of my co-workers to use delicious for his bookmarks. Until then, when we wanted to inform the other about an important or interesting link, we sent an email, saying something like “Look at this” and added the link in the actual mail. This had the drawback that we most often didn’t visit the linked website because we forgot about the mail at some time and that we thought that there is no need to keep the link somewhere, because it’s already in our postbox (thanks to Xobni we were even able to find the links again).
Delicious really eased the thing and made it possible for us to quickly inform the other about new maritime products or interesting news, keeping the bookmark in a well-defined place at the same time. And now comes the interesting question: What should we do with documents from within our company? Of course there is the possibility to also store local paths within delicious (and make them private, so nobody else can see them), but the interface is of course not made for this (copying paths around is really not the thing I want to do).
Additionally, what about meta-information we want to store. Descriptions, comments and guidelines which have no place in a normal document, because they are either updated on a regular basis, should be accessible for everybody within our company and/or where maintaining a word document is not really appropriate and useful. Yeah, I know, the best thing would be a Wiki, and we actually have one within our company. But unfortunately it’s a Sharepoint wiki, which sucks completely. For sure, it integrates into our IT infrastructure just fine, but editing and maintaining your pages is horrible. Everybody can format their text as they like (I haven’t seen two pages with the same font, text size or color), attachment handling is even worse (scrolling through dozens of pages to get to your directory – fun!) and once you’ve pasted a linked to your page, you cannot edit the location, you can only delete it and define a new one.
Our current solution is to maintain an offsite-wiki which works for us pretty fine but has the disadvantage that it’s not integrated into our IT infrastructure, so for example users need to be added manually.
What’s your company’s solution to this problem? Where and how to you maintain information, bookmarks and documents? Is your company even aware of the fact, that it looses traction and money with non-existing information and knowledge management?
Blogging was apparently not the most important thing for me in the last year. Amazing four posts in 2008 and hardly any visitors to this site (except some people looking for Time Machine notifications with Growl – thanks for that!).
New year is just around the corner and traditionally it’s now the time to make resolutions for the new year. Well, here is one of mine: I will write more blog posts in 2009 and make this place a more vivid one. Part of this “initiative” will be some information about my current job and what I do (as far as I’m allowed to tell you about), some geekish – I recently got an iPhone and started developing Cocoa applications – and not so geekish things – purchasing a sewing machine and constructing some small furniture are other resolutions for 2010.
So, the first step into the right direction was changing my old blog theme (Tarski) to something new (Grid Focus by Derek Punsalan), spicing it up with some Flickr and delicious badges, and, of course, adding some fancy social media icons. I also updated WordPress to the newest version and noticed that installing plugins and themes is now a lot easier than it was last year, thanks to the plugin and theme directory. No need to ftp’ or ssh’ to your server, just click and install. Nice.
ESRI released a beta version of their ArcGIS API for Silverlight/WPF some time ago. The may version does not support WMS out of the box but posted some code on their Code Gallery.
Unfortunately this code will not work for WPF (and I doubt that it will even work for Silverlight, considering the strange GetCapabilities parsing and the missing formatter information when parsing floating values). Anyway, I had some time at work and fixed the code for WMS version 1.1.1 and 1.3.0 (I didn’t test other versions, but will do so along the way to some fancy, internal prototype).
Have a look at the code on Github.
But then I turned on the radio and Sunny Side Up was on, and everything was fine again. (I noticed only later that Sunny Side Up is part of the program on Sundays and Holidays).
I got something like 400 unread news items in my Google Reader account, which is quite a horrible number. The easiest thing would be to delete some news feeds but actually I’ve already reduced the number to something like 20.
Some sites, like Apartment Therapy or Unplugged publish dozens of stories per day. Usually I quickly skip through them, star the most important posts for later reading (haha) and add the really, really great posts to my delicious account.
But: Couldn’t Google, the great data and information cruncher, tell me which items are of greater interest to me than others? I starred some 100 items so it should be fairly easy for them to find a general pattern. Would you do that for me, Google?
