Greener electronics

Guide to greener electronics

I knew there was a reason that I liked Nokia and Lenovo so much: they’re greener than almost every other big-name electronics firms out there. But they still have a way to go.

I picked up this story back in April on the PC Pro website: Lenovo out in front in green race. It’s sat in my “to blog” folder ever since.

How green is your gadget?

It was referring to an electronics guide from Greenpeace where they assigned points (out of ten) to the major mobile and PC manufacturers based on their global policies and practices on eliminating harmful chemicals and on taking responsibility for their products once they are discarded by consumers.

In August 2006 Lenovo were sitting at the bottom of the league in a very sorry 14th place (of 14). Within seven months however, the Chinese company who bought out the PC-manufacturing arm of IBM, had managed to completely turn around their green credentials and were leading with 8/10.

From December 2006 to March 2007 the advertisers’ favourite Apple were bottom of the league on only 3/10. By June 2007 the situation was a little different: Apple had moved to a little over 5/10, making Sony the worst — having not moved at 4/10. While Lenovo had slid to about 7.5, with Nokia overtaking at 8/10.

Top 14

The standings as of June 2007 sit at:

  1. Nokia (8)
  2. Dell (7.3)
  3. Lenovo (7.3)
  4. Sony Ericsson (7)
  5. Samsung (6.7)
  6. Motorola (6.7)
  7. Toshiba (6)
  8. Fujitsu-Siemens (6)
  9. Acer (5.7)
  10. Apple (5.3)
  11. HP (5.3)
  12. Panasonic (5)
  13. LGE (4.3)
  14. Sony (4)

Something certainly to bear in mind when choosing a new PC, laptop, mobile phone or other shiny gadget.

You can read Greenpeace’s Guide to Green Electronics online.

My new conceptual model of FeedDemon

FeedDemon

Last night my conceptual model of how my favourite RSS reader, FeedDemon, works when synchronizing with NewsGator online changed.

Conceptual models

In Donald A. Norman’s book The Design of Everyday Things he writes about the importance of conceptual models:

A good conceptual model allows us to predict the effects of our actions. Without a good model we operate by rote, blindly; we do operations as we were told to do them; we can’t fully appreciate why, what effects to expect, or what to do it things go wrong.

As long as things work properly, we can manage. When things go wrong, however, or when we come upon a novel situation, then we need a deeper understanding, a good model.

(Donald A. Norman, The Design of Everyday Things, pp.13-14)

So a conceptual model is just the picture we have in our heads about how we think something works.

FeedDemon RSS reader

My RSS reader of choice for the last few years has been FeedDemon, now at version 2.5.

One of the newest features that FeedDemon offers is the ability to synchronize feeds with NewsGator, an online RSS application. This is really useful if you frequently work from more than one PC (e.g. home, work and laptop) as you can add/edit/delete your feeds in one place, and see the changes reflected in your other locations.

The problem I had

So last night I wrote a long blog post about Aunt Mary’s funeral. I published it to my server, and then checked it in FeedDemon. (I do that sometimes just to make sure that my RSS feed is behaving.)

Nothing. No new posts since Friday. Hmm … so I did a refresh. No new posts. So I went online and checked it on NewsGator itself. Again, nothing.

My good friend Solo Bass Steve was online, so he checked his RSS reader … it was fine: the latest post had published out. The problem I reckoned was then with my copy of FeedDemon.

So I tried the following (this is like a blog montage to save time!):

  1. Installed RSS Bandit and checked the feed there — it was fine
  2. Booted up my laptop and tried FeedDemon there — same problem: only old posts
  3. Deleted the RSS cache in FeedDemon, rebooted, resynchronized — didn’t fix things
  4. Uninstalled FeedDemon, resynchronized — no joy!
  5. Removed my blogs folder from synchronizing with NewsGator – refreshed the feeds: BINGO!
  6. At the same time I checked it on my laptop (which was still synchronized) and the new posts suddenly appeared. I checked the clock and it had just passed the hour.

Things were becoming a little clearer.

What I thought happened

My conceptual model of how FeedDemon works in synchronize-with-NewsGator mode was this: I thought that FeedDemon simply sent NewsGator a list of all the feeds that I’m subscribed to and then downloaded the various posts itself.

I regarded the online version as essentially a master list of all my subscribed RSS feeds, which I could access from the three PCs I regularly work from.

I thought that when I started FeedDemon it check its own list of feeds against those on the master list, update the list as appropriate and then allow FeedDemon to visit each of my 100+ subscribed-to websites and download the latest posts.

That’s what FeedDemon does in standalone mode: it downloads the feeds as-and-when, either on a predetermined schedule or manually when prompted.

As a diagram it might look something like this:

Diagram of PC connecting to various servers

My PC is in the middle, synchronizing the list of feeds with NewsGator on the left, and then on the right pulling in the feeds from my subscribed sites.

What I now think what happens

But based on my tinkerings last night it would appear that not only does FeedDemon simply synchronize the list with NewsGator it also pulls in the latest feeds from there too.

It would appear that NewsGator only updates its feeds at a predetermined interval (e.g. once each hour) — which is fair enough for a shared, online service — and it is based on that last automated check which posts FeedDemon actually pulls in, using NewsGator as a proxy.

So, for example, if NewsGator checks for new posts at 10:00 pm, and someone publishes a new blog post at 10:05 pm NewsGator will not pull that in until after 11:00 pm.

In diagram form it might look like this:

PC synchronizing with servers

My PC on the left connects to NewsGator’s server and pulls in the posts that it has already downloaded on its last scheduled check.

That would explain why updating the feed to my blog didn’t pull in the latest posts: all I was asking FeedDemon to do was to reconnect with NewsGator’s servers and check whether it had pulled in anything new. I wasn’t actually checking that feed itself.

Conclusion

For the most part, unless it’s a frequently-changing website (such as BBC News) I guess most folks won’t need an RSS feed reader that checks any more often than once every hour. So in that sense I can understand why the synchronization-mode has been setup like that

It also saves the NewsGator server having to work unnecessarily hard pulling in data that is only going to be accessed intermittently. That’s the compromise that has to be made, I guess, in offering an online synchronized service like this: you can have synchronized feeds but at the cost of them being at most 59 minutes out of date.

However, there seems to be no way to either manually update NewsGator’s list (by logging into your account and requesting a manual update) or change how often it checks for new posts.

I just wish that this had all been made clear: when you change from the standalone mode to synchronized mode you’re now accessing your feeds via the NewsGator proxy rather than the live feeds themselves.

(Diagrams produced in Microsoft Visio 2003.)

Testing websites in multiple browsers

Browsers

In the words of The Fast Show: This week I are been mostly testing websites in different browsers.

Using Google Analytics

Thanks to our use of the ever-useful Google Analytics I can see that the top five browsers to visit the University website are:

  1. Internet Explorer – 74.01%
  2. Firefox – 17.78%
  3. Safari – 6.65%
  4. Netscape – 0.76%
  5. Opera – 0.44%

Within the Google Analytics stats I can also break down those generalisations and learn, for example, that of those 271,685 visitors who used Internet Explorer to visit the website last month 60% used IE6, 38% used IE7. This gives us an idea of the browsers that we should definitely be supporting.

We made a decision at the start of the coding project to only support IE6/Windows and above, and not to support IE/Mac at all — after all, Microsoft no longer support IE on the Mac so why should we?

The Google Analytics statistics have, thankfully, shown us that we were right not to put a huge amount of time and effort into trying to make sure the CSS code worked perfectly with IE5.5 and below. Of the visitors who use Internet Explorer 99.08% use either IE6 or IE7; 0.92% use older ‘unsupported’ versions of IE.

Testing, testing, 1, 2, 3 … 18

So, yesterday I created a mindmap showing us what browsers were most used and what outstanding problems were still being experienced by these.

To do this I got to install quite a few of these browsers onto my work PC. I now have no fewer than 18:

  1. Firefox 1.0.8
  2. Firefox 1.5.0.10
  3. Firefox 2.0.0.4
  4. Internet Explorer 3.0
  5. Internet Explorer 4.01
  6. Internet Explorer 5.01
  7. Internet Explorer 5.55
  8. Internet Explorer 6.0
  9. Internet Explorer 7.0
  10. Netscape 4.8
  11. Netscape 7.1
  12. Netscape 7.2
  13. Netscape 8.1.3
  14. Netscape 9.0b1
  15. Opera 7.11
  16. Opera 8.5
  17. Opera 9.21
  18. Safari 3.1

This has been incredibly valuable. And I’ve now got the go-ahead to get our hands on a Mac and a Linux box so that we can test their browsers natively.

Firefox

Whenever I’ve tried to install more than one version of Mozilla Firefox I’ve run into troubles; or if I’ve installed the latest release of Firefox 3.

So here’s what I do: I head over to PortableApps.com who offer stand-alone versions of a lot of software, including Mozilla Firefox, Portable Edition.

These are primarily designed to install onto and run from portable flash drives, but they can also be installed to particular folders on your PC, with the reassurance that they will not interfere with your default installation of Firefox.

Specific legacy versions of Firefox can also be found on the Mozilla Firefox, Portable Edition Sourceforge page.

Internet Explorer

Since Internet Explorer is so embedded into your operating system (I’m speaking to Windows-users here) you can’t simply install more than one version of IE quite as easily as you might hope.

However, aware of this the kind people at TredoSoft have created an application that will install five standalone versions of IE: 3.0, 4.01, 5.01, 5.55 and 6.0.

Check out Multiple IE.

Netscape and Opera

With both Netscape and Opera you can install as many versions as you wish — assuming that you remember when installing to give them unique installation locations. Otherwise you can, as I did yesterday, install one version over the top of another.

A great place to download older versions of these, and other browsers is the Browser Archive at evolt.org.

Safari

And last, but not least, Apple Safari 3, which has only been available for Windows’ users for a couple of weeks. It’s most certainly a welcome addition to my installed browser arsenal.

… and lastly

For all you metalheads, why not install the SlipKnot browser from 1999? Not because it’ll help you with your web development, but simply so that you can say that you have a browser installed with the same name as a nine-piece metal band from Des Moines, Iowa! \|m|

O2 Xda Orbit … nearly a review

O2 XDA Orbit - it's a PDA, a phone, a GPS, a Radio ... everything!

It was my intention this evening to write a review of my O2 Xda Orbit, the mobile/cell phone / PDA / GPS / radio device that I got about a month ago.

But instead I’m spending my evening performing a hard reset and reinstalling everything, for the second time this month. And I’m not even going to say how many times I’ve had to perform a soft reset. A day.

That, I guess, is a review in itself.

Error’d: Psion synchronization dialog

Ah, yes. Here’s my favourite user-unfriendly Windows dialog.

This is what you get when you try to synchronize Microsoft Outlook with a Psion 5mx using Psion’s very own PsiWin 2.3.3 and have deleted quite a few of the entries before synchronization:

Synchronizer dialog box

For those of you who can’t read tiny, compressed images of text, it says:

Synchronizer

The Synchronizer has detected 63 missing or deleted items in the Psion. Do you wish to continue and delete the corresponding items?

Click No to retain the items on the other machine.

Click Yes to delete the items on the other machine.

Click Cancel to stop synchronization.

Note: If you have deleted the same item on both machines, it cannot be replaced.

There are a number of reasons that I consider this a terrible dialog box:

  1. Don’t make me think!

    I cannot tell at a glance what I’m supposed to do, without having to read all the text and then work out what on earth it all means. In other words, it’s not intuitive.

    (Following Mike’s comment) What I want is a dialog that I can look at and immediately understand what is being asked of me. I can then spend my time valuably deciding on whether I want to keep that potentially-important data or not. Rather than spending valuable time simply trying to comprehend the text on the dialog box!

  2. Too much text

    Closely related to the previous point: there is too much text. Images would have really helped here; images with the number of missing/deleted items beneath it, perhaps?

  3. Badly labelled buttons

    The text tells me to “Click No to retain the items”, “Click Yes to delete” or “Click Cancel to stop the synchronization”. Why not just label the buttons: Keep items, Delete items, and Stop?

  4. Which machine?!

    The first instruction in the dialog says “Click No to retain the items on the other machine”.

    Which machine?!

    Every time I encounter this dialog I have to stop and work it out, and it always takes me ages: okay, so there are 63 items missing or deleted on the Psion, so the “other machine” must be the PC … right? … right??! So do I want them also to be deleted on the PC? Why could they not just have said: “Click No to keep the items on the PC”?

Confirmation

Once you get past that dialog and decide that yes, you do indeed want to delete the items permanently on both PC and Psion you’re then given the option to back out:

Psion confirmation dialog 2

Confirmation

About to permanently remove items from both PC and Psion.

Are you sure?

The options now, at least, are a more intuitive yes or no. It’s just a shame that you have to practically melt your brain answering the previous question to get there!

Sadly PsiWin is no longer in development — version 2.3.3 (build 149) came out in 2001, and still works with Windows XP, and up-to-and-including Office 2007 — so there is no opportunity to campaign to improve these dialog boxes.

Unless someone is handy with a hex editor … anyone?

Update

Following Mike’s helpful comment below, which made me explain myself a little better I’ve mocked up the following dialogs using Microsoft Visio:

Mock-up of Psion sync dialog box

I have created two here, which (I hope) makes it clear which machines are being referred to and what to do. At a quick glance I can tell on which machine the data is missing or deleted and on which machine’s data I’m being asked to decide. The buttons are also better labelled.

(Error’d entries on this blog are named after the popular Worse Than Failure feature.)