From Delicious to Diigo

2011-03-28 SNAG-00This is the next in my series on online tool migrations a trend affecting us all as Web 2.0 services come and go. This is a hard migration both technically and emotionally as Delicious has been my mainstay social bookmarking tool for many years, since 2003 and the days, and has figured in a good many of my posts. I even used Delicious as my browser home page for 2 or 3 years.

From the technical viewpoint the Delicious service has an extensive list of features that have grown over the years. While the UX has been redesigned significantly on a number of occasions I find it now suffers from an undue complexity even though Delicious more than satisfies my bookmark repository needs.

My own collection of bookmarks rose to about 1100 then, after a severe weeding, a more useful 500 or so. Despite my efforts to simplify my tag count reached 600 which by some effort I cut back by about 50%. Apart from personal use I have used Delicious extensively for holding bookmarks for a good many of my classes. I thought I was tied to Delicious with no prospect of change.

When Yahoo insisted on the use of a Yahoo account for Delicious this introduced an ongoing irritation that still persists. It was strangely apparent that Yahoo was putting little or no effort into taking Delicious forward. Furthermore the Yahoo announcement that the future of Delicious hung under a cloud added to my nervousness. Then hearing continually of the Diigo service from several other higher ed teachers finally made me consider a serious change (I had dabbled with Diigo months before and already had an account).

2011-03-29 SNAG-00

It was the Diigo group facility that seemed the main attractant for my colleagues. From the Diigo graphic you can see groups came in version 3.0 and the last two iterations added features bringing music to the ears of higher ed instructors. Another major stimulus was the Diigo webslides feature which uses Diigo lists discovered by colleague Peta Hopkins. I subsequently used webslides extensively for teaching. (As is the way of Web 2.0 Diigo webslides is now replaced for me by Pearltrees in a further migration.)

Of course the most significant migration effort involves the transfer of the bookmark data to Diigo from Delicious. As you would expect Diigo makes this very straightforward by importing from Delicious export link files. If you can’t bear to abandon your Delicious bookmarks Diigo has a feature to save all harvested Diigo bookmarks to Delicious as well.

First though I needed to spend a lot of time cleansing my Delicious bookmarks. I found many of my older bookmarks were redundant, superceded or had become irrelevant with age. I determined to quarantine all my bookmarks used for teaching. For example the bookmarks contributed by students in my September 2006 class about web applications still exists as This URL will be forever branded in our Blackboard learning management system amongst the online materials for that class.

Eventually I reduced my Delicious bookmarks from about 800 to 102, of which 93 are quarantined from past classes. My surprising discovery was that I had only 9 bookmarks likely to be of use in the future and worth transferring to Diigo! Equally disappointing is my remaining tag list contains 78 different tags!

This migration experience has been invaluable in that it shows I need a stricter, more disciplined approach to bookmark storage. Both the Delicious and Diigo browser integrations make it trivially easy to add bookmarks. It is clear from this exercise that I need more strength of will and bear in mind:

  • the bookmark collection needs constant maintenance, probably once a month at least
  • keep the number of tags as low and as general as possible, and treat solo tags with high suspicion
  • bookmarks that age quickly should not be stored and located with just-in-time search instead, but provided always that no special proper names or obscure keywords are involved – mapping these to easily remembered tags is preferable
  • Even in a few months my Diigo collection has risen to 275 bookmarks and I’m guessing less than 20% will be used again. Perhaps the golden days of bookmark repositories are coming to an end.


Cloud Drive and Cloud Player Excite

Amazon burst out of the blocks this week with the announcement of Cloud Drive and Cloud Player, aptly named cloud computing services. The more generic Cloud Drive gives a free 5GB of cloud storage (leverages the S3 service) intended to store your digital media but can also store any type of file. Cloud Drive can be used independently of Cloud Player which allows you via a browser on Windows, Mac or Android to play back your media, anywhere, anytime – online backup and playback for your media in one, all included in your Amazon account. The ZDNet article describes how this might raise copyright issues, but we shall see.

One report I read indicated these were US-only services but they worked well with my Amazon account from Australia. It was really simple to get started:

2011-03-30 SNAG-00 2011-03-30 SNAG-01

Of course Amazon now offer to deposit a copy of any media purchased from them into your Cloud Drive, but as of the time of writing we are not able to purchase media from Amazon so only our own media can be uploaded. To control the uploading process Amazon provides the MP3 Uploader desktop app to synchronise your media from Windows Media Player and iTunes (AAC is supported but not WMA). I have only about 800 songs in the supported formats so at 3.6 GB they fit well within the 5 GB limit. I am still uploaded and it appears I will need 10 hours due to my 1 Mbps cap on the upload on my Internet service.

This is a very significant advance by Amazon and is just the service I have been waiting for. When Cloud Player works on my iPhone my cup will be full.

More Nails in the Coffin of the Digital Native Concept

My interest was sparked by a post by Mark Bullen about the digital literacy of ‘digital natives’. Mark’s post certainly lives up to his blog title, ‘Net Gen Skeptic’. He points out that research is increasingly discovering that

"digital natives", in fact, appear to have a superficial understanding of the new technologies, use the new technologies for very limited and specific purposes, and have superficial information-seeking and analysis skills

I observe this situation continually with students in all my classes, particularly for those just starting at my institution.

Mark goes on to compare the optimistic views of Tapscott, Palfrey et al, with the peer-reviewed findings of Hargittai and colleagues to show the large contrast in views. Fascinating reading.

Despite covering the features in classes and tutorials it still amazes me how little the ‘digital natives’ use online information management sites like SkyDrive and Google Docs. It is even more puzzling that note-taking sites like Evernote and OneNote are not central to the everyday studies of all students. I would have given my eye teeth for such tools when I was a student.

Liking Windows Live Sync Beta, Mostly

My MeshSince Live Mesh, more formally known as the Mesh Operating Environment,  launched I have been a dedicated user. With the free 5 GB of online storage space I have synced and backed up 9 of my machines, both physical and virtual, with my key working set of files. These files are my teaching materials, research project files and some administration information. Whenever I acquired a new machine, I simply installed the free Live Mesh client and within a few hours (Mesh is not fast) all my current key files were available there. Any change to any file on any machine is immediately synced to the web store and all machines. Moreover on any Internet-connected machine I can upload and download any file. It has always been a puzzle to me why everyone is not using Mesh as Macs are supported as well. Backup and sync with Mesh are perfectly sorted.

imageBut now we have the phase out of Mesh to be replaced by Windows Live Sync and Windows Live Devices which are now fully integrated into the free Windows Live Essentials wave 4. The public beta of Live Essentials which includes Live Sync is now available. Since Mesh has been so reliable I decided to switch to Live Sync Beta over the last few days.

Installing Live Sync warns you that it will replace Mesh and require you to re-sync with the web store, fair enough. I synced up my files from one machine to the new web store without a problem. Then I expected Live Sync would realise the files on all other machines would be identical and require no syncing. No such luck. All the files on all machines need resyncing – a real negative, even though this process is completely automatic.

Another limitation of Live Sync is that its free web store is now reduced to 2 GB from 5 GB, I guess to come in line with other free online sync services. Fortunately I had been careful in my Mesh file collection which had never exceeded 1.7 GB. Thus this new limit has not effected me, at least for the moment. A positive is that the Live Sync web store is now completely integrated into your free 25 GB SkyDrive used for Office Live, photos, videos and sharing stuff.

Live Sync MeshDespite the resyncing I was soon up and running but then hit the next snag. My ancient office machine is still running Windows XP which cannot support the new Live Sync, presumably a less than subtle hint to upgrade to Vista or 7. However, my home desktops, a several laptops and netbooks are fine and are all connected quickly to my new Live Sync ‘mesh’.

Not only was Mesh fairly slow in its syncing but it was a notorious processor hog. For my machines at least, Live Sync uses considerably less processor and seems much quicker to sync. The visual feedback and progress bars of Mesh, though, have not carried over in full to Live Sync. As yet we get only terse indications of sync checking and file transfers in the new Live Sync dashboard shown in the last image.

It is good to see the remote connections features of Mesh carry over to Live Sync. Provided it is enabled on the remote machine it is possible to connect, login and use the other remote machines in your mesh. The connection experience seems smoother and more responsive.

So in keeping with my academic assessment procedures I would award Live Sync an 85% so far, and hope for more marks in future as we move out of beta.

A certain social media guru and librarian of my acquaintance also resident on the Gold Coast would probably kill to have been using these sync services when her laptop went missing this week.

Hoping HyperWebCard Could Regain the HyperCard Glory Days

This post is a ‘call to arms’ for developers to rally round and help create the HyperWebCard environment to allow a wide audience to generate powerful database applications that were possible in the glory days of HyperCard.

I would go so far as to say that when Apple’s HyperCard appeared in 1987 it was a disruptive innovation. It should be added I was a Mac fanatic from 1984 to the release of Windows 95. HyperCard is a rapid application development (RAD) environment that ‘combines database capabilities with a graphical, flexible, user-modifiable interface’ based on the metaphor of index cards with dynamic, interactive content. Developing for Mac OS 9 was notoriously hard (whinging OSX and iOS4 developers please note). So the convenient dynamic language programming offered by built-in HyperTalk scripting and the superb GUI development environment of HyperCard was hugely attractive to programmers and non-programmers alike.

A common phrase used for HyperCard was ‘programming for the rest of us’. Sophisticated, responsive Mac apps with local databases could be built with HyperCard even though HyperTalk was interpreted. My HyperCard bible became the HyperProgramming book by George Coulouris and Harold Thimbleby.

HyperCard has a small set of integrated components:

  1. A local database of card stacks with a designated home stack; a stack is like a table in a database, and each card is equivalent to a table row or record; unlike a database a card can inherit content from a background card
  2. User interface elements such as buttons, text boxes, dropdown lists, menus and so on
  3. A object model based on stacks, cards and user interface elements placed on the cards with an associated event model
  4. A loosely typed scripting language, HyperTalk, for writing event handlers and initialisation actions tied to the object model
  5. A run-time environment with display mode (app execution) and edit mode (integrated development environment)

In the 1989-1991 timescale I used HyperCard to implement a Mac application for email called BRUITmail with HyperCard. BRUIT stood for Bond Research in User Interface Technologies and BRUITmail provided the HyperCard equivalent to Gmail, ie a graphical user interface to the command-based email service at Bond at the time. Sadly the source for BRUITmail disappeared with my Macs when I switched to Windows 95 in 1995.

So why this post? Well it started at BarCamp Brisbane V while I was stroking my newly arrived iPad during a talk where the speaker held the strong view that we should be writing HTML5 apps for iPhone/iPad not native apps. I asked myself why should we not be able to create HTML5 apps using the iPad itself? This started bells ringing (especially after a later talk by @spidie on the same!) about how easy HyperCard used to be. Surely iPad Safari could host an HTML5 app generation and execution environment?

I have finally managed to find time to put my thoughts into this post after finishing the one on iPad presentations. It should be possible to carry forward the HyperCard philosophy into these days of the web application. For the sake of discussion I refer to this new web-based rapid application development environment as HyperWebCard. It requires little thought to discover that most of the components needed by HyperWebCard are already with us:

  1. HTML5 local database storage, usually implemented with Sqlite, will be more than adequate to store stacks and cards. See Web SQL Database.
  2. jQuery and jQuery UI will provide the user interface elements together with exciting additions
  3. HTML5 DOM is a more than adequate replacement for the HyperCard object model; a card maps directly to a web page, and card contents map to a variety of HTML5 tags. See the HTML5 working draft
  4. JavaScript, the lingua franca of the web, is the natural replacement for HyperTalk
  5. The HyperWebCard run-time environment is the missing component, but surely this is just a relatively simple JavaScript library providing a visualisation of the stack/card database with the two modes of display and edit? Obviously a drag-n-drop interface for creating/editing card contents is needed and JavaScript editing/storage must be supported in the database.

This post is a plea to software develops to take up the call of HyperWebCard as I don’t have the time commitment or expertise to implement this fully. Hopefully the promise of wide applicability across all platforms not just the iPad will provide the inspiration for a team of open source HyperWebCard developers and designers. Won’t the prospect of developing apps on the iPad itself be super attractive? At the time of writing all hyperwebcard.* domains are still available. Who will be first to grab the domain and run with the project?

At this point I had a quick look on stackoverflow for more HyperCard information and found the question by Gabriel Cuvillier aptly entitled ‘Is there a web application equivalent of Hypercard?’ so I am not the only one thinking along these lines. One answer to this question was Google App Engine but this is currently definitely not a RAD.

I surely hope the developer community can recreate the glory days of HyperCard again in the new HyperWebCard project or equivalent.

QR Codes Come for Free with URL Shortener

Via a @timoreilly tweet comes a buzz from Matt Cutts which is shown here for the buzz challenged:

8-04-2010 SNAG-00

So when you use the URL shortener (see my tweet about the Chrome extension) you just add .qr to the URL to see the QR code for that URL. So simple and yet so cool.

This blog is at and the QR code at which should show the QR code for my blog:

8-04-2010 SNAG-01

960 x 1080 – The New Ideal Size for Web Pages

Like me you you are running Windows 7 on your desktop with a shiny new full HD (1920 x 1080) monitor or two or three (I wish).

You see one of the 7-second Windows 7 ads on TV that dragging any application window to the left or right of the screen neatly resizes the window to occupy exactly half of your screen, that’s 960 x 1080.

You instantly love this feature because you now see two windows side-by-side and can so quickly read/switch/copy-and-paste between the two.

You start to use it for everything, Chrome (Gmail, Gcal, Greader), Firefox (Gdocs, Gsites, Remember The Milk), Evernote, Tweetdeck, Windows Live Writer, Skype, iTunes, Windows Explorer, Outlook, Word, Excel, and so on – all neatly side by side and quickly navigated with Windows-Tab. Ecstasy.

side-by-side 7-04-2010 10-28-22 PM

But there’s more. You discover that Windows-Left and -Right Arrow save you having to drag the windows left and right, and used repeatedly cycle between left half, original and right half positions. Wow – Microsoft actually thought of these miracles? Windows- Up Arrow makes the current window fill the screen – superb. Windows-Down Arrow returns to original size and used again minimises to the task bar. What magic!

The upshot of all this – I read all web pages at 960 x 1080 size. But you all ask, isn’t 1024 x 768 regarded as the ‘norm’ for web pages? Sadly this is so and forces me to scroll to right a lot – very frustrating.

So to keep me and my soon-to-be hundreds of millions of fellow Windows 7 users happy, make sure your web pages are usable at 960 x 1080 size. Please!