I’ve been using Bloglines for a while now to keep up to date with RSS feeds and their associated sites, including news, webcomics, blogs and so on. It’s a great web application for organising feeds and keeping track of which articles are unread. It also has the added perk of grabbing the contents of feeds even when I’m offline, ready for me to read when I log in next.

The problem with Bloglines is that it’s buggy, occasionally failing to actually display feeds, and quite resource hungry for the browser such that mobile browsers – even on the powerhouse that is the N900 – struggle with it. It’s also not very finger friendly for the aforementioned device.

Therefore, I decided to put together my own tool that reproduces the functionality I enjoy from Bloglines, while being better suited to a mobile browser – including a certain amount of finger friendliness. It’s not quite finished yet, and the it doesn’t download any content when you’re offline, but I think I should at least offer a link to my project to the outside world.

And so, I present, GaliRSS (yes, it really needs a better name…)

Beyond what I’ve already mentioned, a handful of its features include (in no particular order):

• An AJAX driven interface – almost no reason to reload the page.
• Caching of feed content, so feed content can be read later even after it drops out of the feed itself. The cache is currently limited to up to 4 weeks of history.
• If a feed only offers excerpts of content, the actual content can be either loaded in a new browser window/tab, or in an iframe within GaliRSS – no need to leave the site if you don’t want to.
• Feeds can be organised into groups of similarly themed content, available from a sidebar (which can be collapsed or hidden entirely).
• Feed items can be ‘pinned’ to read later.
• The feed list can either show all unclicked (unread) items, or items from a single feed, or items from a group of feeds, or pinned items.
• A ‘cleanup’ button is provided to hide clicked items. Clicking it a second time will reveal them all again.
• Each feed is colour coded, with a ‘strong’ colour picked at random for each new feed. This colour can be changed easily.
• Usage of GaliRSS is quite customisable from the Settings page.

At present, it doesn’t support authenticated feeds, or feeds with a https:// url (ie. http:// only)

Note that it’s not entirely break-proof at the moment. I also couldn’t completely guarantee the security of it at this stage.

Comments are welcome, and I’ll probably keep working on it for a while yet – but it is provided with no official warranty or support for now.

One thing that I did appreciate about the n900 was that doing a complete flashing of the firmware didn’t wipe things like the text message log, and a lot of application settings survive presumably by being safely stored in the mass storage somewhere. Sadly, the downside of this is that when an application breaks (most notably in my case, the built in Ovi Maps), there’s no obvious way to fix it, including reflashing the firmware. Nokia do, however, offer blank image of for the mass storage that you can ‘flash’ onto the device. Before doing this though, I’d like to take a copy of a few things that I do want to keep…

Anyway, whatever your reason, taking a copy of your SMS / IM / call log is actually pretty simple. All it needs is a command to be executed within a terminal window.

Before doing this, I would recommend installing an SSH server (available from the Application Manager) and using a tool like Putty to connect to your device so you can use a normal keyboard and/or cut and paste from here…

Start out by opening a terminal and using cd to find a directory you want to put the backup in – the default location is probably fine if you’re using a terminal on your phone. Otherwise, you might want to do something like cd /home/user/MyDocs/ before doing anything else.

The following command, tweaked slightly from here, will dump all of the data from your phone’s communication database into a file called output.html:

sqlite3 -html /home/user/.rtcom-eventlogger/el.db “SELECT start_time,service_id,event_type_id,free_text, remote_uid FROM Events ORDER BY start_time DESC;” > output.html

This will create a file called output.html, which you can copy to your PC. For some browsers to view the file properly, you may need to open it in a text editor and add <html><table> to the top and </table></html> to the end – the command seems to omit these parts.

Once you’ve got the file open in a browser, a couple of the columns could benefit from some explanation. They’re displayed in the order they were mentioned in the command (there are no headings in the file):

• start_time: The time the item ‘began’, I presume. For text messages/IMs, this would be the time it arrived/was sent, whereas for calls it represents the beginning of the call.
• service_id: 1 for a call, 2 for an IM, 3 for SMS.
• event_type: This value depends on the service_id – there are two values for each. The larger value represents sent (outgoing) items, and the smaller represents received (incoming) items. For instance, for SMS (service_id is 3), the event_types for sent and received are 6 and 5 respectively.
• free_text: The actual text content of the message. Blank for calls.
• remote_uid: For calls and SMS, this is the phone number of the remote phone. For IM, this is the email address or screenname of the other half of the conversation.

Of course, there’s no easy way to restore the messages to the database afterwards – though some clever coding could probably pull it off. There is a Backup/Restore feature built into the phone if you want to go down this route, but personally I’m loosely planning a complete wipe and would like a stash of my old content stored somewhere, just in case there’s something I want to come back to later.

I recently started keeping up with Leo Laporte’s TWIT podcasts, with its very American technology content with accompanying chatter. While it is very much put together from a US of A perspective, they do make nods to us over here in Europe where its relevant, and he and his guests genuinely seem to be having a good time. Anyway, they have a habit of mentioning technologies, software and links that inspire my curiousity. Since I’m often out and about while listening, and don’t want to be distracted from the podcast too much, I note things down on my trusty N900 to look up later. Given this usually inspires only 2 or 3 items per podcast, I suspect that these may be a source of many posts on here in the future, given my current track record of posting at least. It’s certainly more manageable than trying to put up content covering everything I think is cool on Engadget or Gizmodo…

The first, small item I want to mention is gdgt.com, which seems like a pretty cool idea, at least for gadget lovers out there. It’s also worth noting that it’s the brainchild of the previous founders of the two major aforementioned technology blogs…

Gdgt (GDGT? gdgt?) is something of a social network for people and their gadgets, and that’s about it. You’ve got your normal list of friends and a profile as with any other social site, and it’ll let you hook into big names like Facebook, Twitter and Gmail to find more friends automatically. What it then does is give you the ability to add gadgets to your profile, under the categories of “Have”, “Had” or “Want” to represent your relationship to this gadget. Personally, I somewhat would have liked a couple more categories – notably “Used” and “Have Access To” to represent those I’ve played with, or those that are owned by family/housemates living in the same building – or that are available to me in my workplace. That said, I suspect they were aiming for the tried and tested intention to keep things simple.

The site then goes to offer you the chance to discuss and review gadgets (whether they’re linked to your profile or not) and to look up the gadgets your friends have against your name. There’s also a news feed, which seems to display news about random gadgets, and a ‘features’ page which seems to contain blog posts and news updates from those behind the site itself.

The site very much feels like it’s just starting up still. The functionality they have in place is solid, and just about unique enough to stand out in the world of social networks. There’s a lot they could do in the future too, which would be great as long as they keep it simple. Ideas that sprung to mind off hand included:

• Finding other users with the same gadget(s) as you? Perhaps the ability to search based on an overlap of a specific number of gadgets, 1 – n (where n is the number you have on your profile). Connecting this up with a location based search would have a lot of potential – I’d love to know how many other N900 or Gigabyte T1028X owners there are in my corner of the world, for instance.
• It would be nice if the news feed could be (optionally) toggled to only show news about gadgets you have on one of your lists (either all of them or specifically). There’s a lot of news about the next iPhone appearing, and to be honest I’m not that interested at the moment.
• Going a step further, the site has some awareness of manufacturers behind devices. It would be cool if the site’s news feed could offer me news specific to them, so I can see what’s coming up in the future. After all, Nokia just announced the N8, and there’s no sign of that on there.
• It’s nice that users can ‘like’ news posts and rate gadgets, and that you can get a view of what’s popular as a result, but it would be nice to develop this on the social side a bit more, giving me more feedback from my friends, or from others in my corner of the world. I imagine they could pull out some really interesting figures about what gadgets are used where as their membership builds up…

Of course, the site is still in beta, so there’s a long way for them to go yet. I just hope that they can get some distance along that road before someone else comes along overtakes them.

In the same vein, I signed up for Twitter to fall in behind other tech-enthusiasts everywhere. I posted occasionally, but tried to keep up with my friends’ activities more often. This was assisted by my Nokia N97 after purchasing a license for Gravity. This made using Twitter an absolute pleasure, and I found myself posting (slightly) more often and glancing at the updates from others throughout the day. However, I then shifted over to my Nokia N900, and its inability to run any Twitter apps even close to Gravity in quality and features…

I tried a number of apps written for the N900 – Mauku and Witter to name a couple. They did an OK job, but generally felt a bit clunky or scruffily presented. It then dawned on me that the N900 is designed to be ‘always connected’ and has a blindingly good web browser built in, so I tried out dabr. This site claims to be a ‘mobile web interface to Twitter’s API’, and it certainly succeeds at that. In the absence of Gravity it would have been a fantastic tool for use on my N96 or N97 in the past. For the N900, with the high resolution screen, the text was just too small and the links were far from finger-friendly.

It was at this point that I started to consider writing my own Twitter client. My development skills for Maemo are far from perfect, and I came to the conclusion I wanted something that I could use on multiple platforms anyway, so something web based seemed to be the way forward. I decided it would have to be finger friendly, but ideally in a way that wasn’t too intrusive when I used it from a desktop browser.

PHP and Javascript are my current languages of choice for web development, so I set to work on decyphering the Twitter API (not a difficult task) and exploring code examples for actually pulling data out of Twitter.

It was around this point where I happened upon TweetGo.net, which already seemed to do the basics of what I was planning – a finger friendly UI for Twitter. However, it’s also in its early stages, and lacks a few features that I’d like to see in a client – which would recreate a lot of the functionality and ease of use of Gravity.

And so, GaliTwit was born. You can check it out in current (also early) form by following this link, or from the link I’ll shortly add to the navigation bar at the top of this site. I’ll hopefully write some more posts about it in due course to cover what features it already supports, and what I plan to implement in the future.