dkeats.com

My Wits blog Tweet
560 days ago

Since I am kind of up to my ears in baby crocodiles at the moment, and I am doing some blogging on the KIM site at Wits, you can access it there, if you want to see what's happening. I will be back  blogging here shortly.

http://kim.wits.ac.za

---

Computer Science for Fun with MXit, Chisimba and Google Talk Tweet
573 days ago

Here are the slides from my presentation at the Google Faculty Summit, where I talked about the Computer Science for Fun project that I initiated after last year's Faculty Summit. This is a project that is still under way at UWC, as part of a wider initiative with schools on the Cape Flats.

Also available from http://chameleon.uwc.ac.za

Unfortunately, I left the microphone for my iPod behind, so I could not podcast it.

---

At Google Faculty Summit in Zurich Tweet
577 days ago

I am at the Google Faculty Summit in Zurich this week. I will Twitter from time to time using the hash chode #faculty_summit. This blog will aggregate those Tweets in one place using search RSS and a Chisimba filter. You will of course need to visit this page from time to time to see the posts, or follow me on Twitter.

[RSS]http://search.twitter.com/search.atom?q=%23faculty_summit[/RSS]

---

Web 2.0, Library 2.0 & stuff X.0: my remarks to the Wits library staff Tweet
589 days ago

This podcast is just some impromptu remarks made to the Wits library when introducing myself as the Deputy Vice Chancellor in whose portfolio the library is located. It was off the top of my head, and included question and answer, so there is nothing prepared or organized in it, so listen to it from that persepective please!

 

---

Charged magazine's disgusting article on "Open Source" Tweet
593 days ago

I popped into CNA in the Kilarney Mall in Johannesburg this morning to get something to read with my coffee and breakfast. Never having bought a copy before, I picked up the January issue of Charged magazine (http://www.charged.co.za/) because it had a cover story on 'Open Source Software' (OSS). I paged through to that article, being curious what the magazine might have to say.
 
Shocked!
 
That's what I was when I read the article. If the standard of journalism in that article is typical for the rest of the magazine, then the magazine is unlikely to contain any useful factual information about anything. The depth of research shown in the article is staggeringly shallow, and the choice of words devious. I will never know, because I will certainly not buy another issue, nor will I bother reading anything else in that issue to test the quality hypothesis.
 
The article could not even come up with a sensible definition of 'Open Source Software', even getting the basic definition wrong. It talks about the 2004 Venezuela decree to adopt OSS, but does not even mention that South Africa has a strategy on Free and Open Source Software (http://www.oss.gov.za/docs/OSS_Strategy_v3.pdf). 
 
The section 'Questions and Answers' starts off with an interesting use of the English language. It then asks whether the push for OSS is a 'leftist leaning towards avoiding commercialized software'. While it is clear that the author is establishing two alternative scenarios, this style of writing is just very poor quality. It reinforces the notion that OSS (or Free Software, or Free and Open Source Software [FOSS] as I prefer) is not commercialized. The fact is that most useful Free Software applications have a commercial basis to them, and not clarifying this perpetuates the non-commercial myth.
 
The article, unintentionally I think, creates the impression that to explore FOSS you have to install the GNU/Linux operating system (incorrectly referred to as Linux). In fact, there are FOSS alternatives to all of the proprietary packages mentioned that work perfectly well on the various Windows operating system versions. Open Office is a an excellent alternative to Word, Excel, PowerPoint and Access; the Gimp is a viable alternative to Photoshop.
 
Recommending that people take their first step into FOSS by installing GNU/Linux is absurd. As a long time FOSS advocate, I would not do this, instead suggesting that people instead check out some of the Windows programmes that are FOSS alternatives to the the commonly used proprietary ones. This way, people can gain experience in a familiar environment before taking the GNU/Linux plunge.
 
The article then provides a damaged description of the 'Linux kernel' and states that 'one of the most popular forms of the Linux kernel is Ubuntu'. Ubuntu uses the Linux kernel (notice I did not say GNU/Linux kernel!) and is not itself a kernel. It is rather a distribution, or distro in common GNU/Linux parlance. This is a vitally important aspect of understanding GNU/Linux, and should be understood by anyone attempting to write a magazine article on the subject.
 
The article makes no mention of the fact that you can install Ubuntu within Windows, so if you are a Windows user, you can install and play with Ubuntu as if it was a Windows application. This is a good way to acquire familiarity before making the switch. Another way to explore Ubuntu is to use a live CD, and although this results in a slow system, it is a way to try before you switch. Lastly, the ability of Ubuntu and other GNU/Linux distros to coexist alongside Windows on the same machine is not mentioned.
 
The section "The Differences" creates the impression that you have to use a terminal to use Ubuntu. As a long time Ubuntu user, I sometimes use a terminal by choice because it is faster than clicking a whole lot of graphical icons, there is almost nothing that a normal user would need to do that would require a terminal.  It fails to understand the amazing user-friendly application Synaptic, instead choosing to use loaded words like "Ubuntu developers have attempted to divert users". It says that programmes not on the default list of packages are problematic to download, something that I find absurd being equally familiar with Windows and GNU/Linux. It is far easier to find and install packages on GNU/Linux than it is on Windows, and the ability to add repositories to Synaptic enables you to install almost anything using its simple graphical user interface. Further more, you do not have to buy them or steal them to install them.  I don't think I use a terminal to install any software, except where I choose to do so because it is faster.
 
The article completely fails to understand the file permissions side of GNU/Linux, treating this strength as a confusing weakness. However, the notion that you need to use a terminal to empty the trash is absurd. It says the much of the time you will find yourself scanning the Ubuntu forums for command line solutions to simple technical problems! I found such a statement utterly absurd. Most of the time I use my computer do do useful work using software that is generally as easy or easier to use than its Windows counterpart. My entire family uses Ubuntu, including my wife who is not a technically savvy user.  If you want to get into deeper things, just like you have the command line on Windows, you have the command line on GNU/Linux. When you get used to it, you can often do work much faster than you can with a GUI, but there is almost nothing that I would do in the normal course of using my computer that requires me to use a terminal. Suggesting so is just plain misinformation.
 
Under the section '"O" for Open Source Apps', the author creates the impression that Adobe Acrobat Reader is not available for GNU/Linux. Again, this is evidence of the shallow, inaccurate nature of the article. Acrobat Reader can be installed from Synaptic by enabling the appropriate repository. 
 
The author further suggest that Adobe Photoshop is better than the Gimp, something that is a highly personal thing. I have used Adobe Photoshop since version 1. I recently switched to the Gimp, and I have come to prefer the Gimp by a long margin. The only thing I miss from Photoshop is 'styles', but that is more than compensated for by other Gimp functionality. So making such blanket statements as though they were absolutes is just plain misleading.
 
The article says "the Ubuntu kernel comes packaged with a mail client", which is rather like saying the Toyota Yaris engine block comes packaged with a back seat. You only realise how silly such a statement is when you are familiar with the subject.
 
In the insert on "The Downsides" it is stated that Ubuntu (still mistakenly referred to as an operating system) does not have some of the user-friendly attributes associated with Microsoft Windows. I would challenge the author to list those missing user-friendly features, because as a user of both Windows and GNU/Linux, I cannot think of any. I can think of lots of user-friendly features in Ubuntu that are not in Windows, but none the other way around. 
 
In the same section, the author says if something goes wrong, you have to go to forums to get information. He sees this as a 'downside', and says that with 'commercial software' (Ubuntu is commercial software, darlings), you have a helpline that you can call to get your problem solved. I have been asking people about this for many years, and I have run a department responsible for some 4000 or more computers, and I have never met someone who has called a support number for assistance. When I was a Windows user, I tried it once, and failed to get any useful response. I ended up solving it by searching user-driven forums. When I have had issues, I have seldom had them for more than a few minutes, almost always finding the answer by Googling or by asking in community forums. This is not a 'downside', it is one of the most awesome positive features of working with Free Software. However, if you want commercial support, you can also get it for Ubuntu, and also have a hotline not to call.
 
The 'downside' links OSS to South American countries, neo-colonial capitalistic considerations, and suggests that you have to be an early adopter to use OSS. This is just plain dishonest, showing the author's iignorance and lack of professional journalistic ethics, but the fact that it is in a magazine that naive computers might just trust is a sad tribute to the editorial and journalistic standards of the magazine.  Shame on you editor, I say shame on you!
 
In disgust, I did a $mv /Charged ~/.Trash followed by a $sudo rm ~/.Trash/* -R. Never again will I buy a magazine that publishes such low quality articles. But thats jut me. You can make up your own mind!

---

Life in a technology backwater Tweet
595 days ago

Thanks to a long history of national IT policies and practices that are designed to protect vested interests and colaterally promote poverty, the internet and IT landscape in South Africa is best described as a total mess. The fact that we have a tiny little IT industry that is pretty good obscures what has been lost since 1994 due to these policies and practices.

We are beginning to emerge from this mess, but we have a long way to go. Attempting to open google.com this morning, I got:

Server Error
The following error occurred:
Could not connect because of networking problems. Contact your system administrator.
Please contact the administrator.

The same error occurs on all website, despite me spending in excess of R1000 a month on miserably poor bandwidth of abismal quality. It is hard to smell the roses when you are floating in an ocean of freshly deposited feces. Its hard to be cutting edge, when you are surrouned by rusty lumps of pig iron.

Hopefully, soccer provides more incentive then knowledge and logic for improving the situation, and that the promises of better and cheaper bandwidth will materialize. Its sad for our nation that it takes World Cup Soccer to spur it on, that the logic of development is inadequate motivation for change!

---

Phycology 2.0: Can we foster a new interest in phycology using the technologies of Web2.0? Tweet
595 days ago

Since the emergence of the concept of Web 2.0, it has become fashionable to add 2.0 after concepts to indicate that social-collaborative processes and technologies have been brought to bear on them. Licenses that permit the sharing of the expression of knowledge in digital and other formats have evolved and become a popular element of Web 2.0 and the digital culture of the 21st Century. The technology-society coevolution that is happening at present may have some potential to rekindle an interest in phycology. As a recognized science, Phycology is in decline, having become almost a subculture or other disciplines during the past 1-2 decades. A scan of current social networking and social media applications show that there is almost no discussion of algae or media about algae available. Popular websites focusing on algae contain no social, collaborative features. Popular microblogging applications, such as Twitter contain no mention of algae by any professional phycologist. Photographs of algae on Flickr are almost all by amateurs, and generally of poor quality. Most online artifacts produced by Phycologists are fully copyrighted, and most phcological research is published in closed access journals that retain full copyright. Relatively few phycologists exploit new scholarly forms of communication, such as blogging, podcasting and presentation sharing.  In this talk I explore these trends, and suggest how organizations such as PSSA might help to use social technologies to help make our science better known.

---

Using Mozilla's Flash plugin for Flock with Ubuntu Hardy Tweet
598 days ago

I have been using the social  browser Flock quite a lot recently on my desktop (Ubuntu / Intrepid), but I packed up my desktop, so I just installed Flock on my laptop (Ubuntu Hardy ). Of course, it did not work with Flash content, so the following command was used to fix it.


Since it took me a  while to figure it out, I am just posting it here in case I need to do it again! You may consider this blog pollution and ignore it!

Tags: flock, flash, hardy

---

The burdon of attribution Tweet
599 days ago

The requirement for attribution in building on or using Creative Commons licensed work is a non trivial matter. I am not sure if this is a reasonable way to do it, but I have decided to maintain an attribution file for all sources that I use in presentations. The file can be downloaded here ([ALERT:url=http://www.dkeats.com/usrfiles/users/1563080430/attribution/attrib.txt]view[/ALERT]), and I will endeavour to keep it up to date. It is just about impossible to maintain a file for individual presentations since I mix and match between presentations. We really need tools for embedding attribution in the assets and persisting them even when the assets are repurposed. This is technically non-trivial. Hopefully, my file will serve to keep me legal.

---

New release of Chisimba out Tweet
601 days ago

The next release of the Chisimba PHP5 framework (Chisimba-2.1.0) is now available. Chisimba is the application that is used to power this site, and it is pretty awesome, being able to be almost any Web 2.0 enabled system you can imagine from eLearning (Learning Management System) platform, to blog, to content management system, to MXit interface, presentation sharing, to realtime presentation tools, to wiki, etc. With lots of filters for external content, and running in the Amazon Cloud it is infinitely scalable.

Major enhancements included in this release are:

 - Numerous enhancements to the database abstraction layer for increased
performance
 - Numerous core bugfixes and enhancements
 - Patch descriptions added in module catalogue
 - Layout and skin enhancements
 - Increased security and RC4 encryption of session data
 - Complete authentication system overhaul
 - "Remember me" functionality added
 - URL rewriting
 - Remote popularity contest module
 - Additional filters for rich content
 - Some installer fixes
 
and, of course, new modules to add onto your installation!

Please take a look, download it and give it a test drive!
 
Chisimba, for those that don't know it already, is a PHP5 framework made
in Africa, for Africa. It is a collaboration between around 16 African
Universities, as well as around 35 active developers from around the
continent.
 
It can be downloaded from AVOIR at:
 
http://avoir.uwc.ac.za/

and the documentation can be found at:
 
http://avoir.uwc.ac.za/index.php?module=news&action=viewcategory&id=gen14Srv6Nme27_7167_1219410313

There are server setup instructions, as well as installation
walkthroughs available linking from the main AVOIR site:
 
http://avoir.uwc.ac.za/index.php?module=news&action=viewcategory&id=gen14Srv6Nme27_2077_1219410069
 
For those interested in developing a module, or just getting some
additional info please take a look at:
 
http://avoir.uwc.ac.za/index.php?module=news&action=viewcategory&id=gen14Srv6Nme27_6705_1226737050

---

Creating a presentation repository: capturing knowledge otherwise lost Tweet
603 days ago

Every year, in Universities all over the world, hundreds of thousands of conference talks are given, most of them based on presentation technologies such as PowerPoint, KeyNote, Open Office Impress and others. Seldom do these presentations find their way into institutional repositories, and most institutional repository tools are not designed to accommodate them in any kind of useful way. There are sites, such as SlideShare and Scribd, that allow you to post presentations online, but they are not institution based. However, they do offer some 21st Century features that repository software does not generally offer as far as presentations and collaboration potential is concerned.

We have the basis for building presentation repository already completed and ready for use. In what started off as a small project with San Jose State University in California over a year ago, we built a presentation sharing facility on the Chisimba framework as a means to develop further semantic technologies for engaging with presentations. The machine learning work has not been done, but the technology has evolved into an excellent system for sharing presentations in a Web 2.0 presentation repository manner. You can view the UWC installation of it at http://chameleon.uwc.ac.za, although it is not being used at UWC as an official institutional presentation repository. The Chisimba module that provides this functionality is called WebPresent, and it integrates with a set of realtime tools that provide realtime presentation capability over the web.

[ALERT: url=http://www.dkeats.com/usrfiles/users/1563080430/repositories/wpres1.jpg][/ALERT]

Chisimba is a Web 2.0 distributed application framework developed in the African Virtual Open Initiatives and Resources (AVOIR) project. Being distributed, having an open API, and being extensible, it is quick to plug in new funcitonality from other applications.

[ALERT: url=http://www.dkeats.com/usrfiles/users/1563080430/repositories/wPres2.jpg][/ALERT]

Chisimba's WebPresent module allows users to upload presentations, license them, tag them, blog them, and build a community around them. They are also inter-converted into other formats automatically, and can be downloaded in those formats, including PowerPoint, Open Document, Flash and PDF. Dublin core metadata is included, as are Creative Commons licenses. Filter widgets and code snippets are provided that allow the presentation to be embedded in any webpage that supports Chisimba filters or allows the use of HTML snippets. There is even a Chisimba filter and code snippet that allows you to embed a live presentation into a blog or other webpage.

[ALERT: url=http://www.dkeats.com/usrfiles/users/1563080430/repositories/wPres3.jpg][/ALERT]

It would be interesting for an institution such as Wits or UWC to build an institutional presentation repository around this tool. It might start to create a sense of what a 21st Century repository should do, as well as provide some basis for promoting this often neglected form of research output.

[ALERT: url=http://www.dkeats.com/usrfiles/users/1563080430/repositories/wPres4.jpg][/ALERT]

Like all software that have produced in ICS at UWC and in the AVOIR project, the tool is Free Software (open source) and available for anyone to download and use to set up an institutional repository. It incorporates a number of technologies that make it scalable from a few hundred to a few million users, given suitable infrastructure on which to run it. It also works in the Amazon Cloud, and other cloud computing technologies.

[ALERT: url=http://www.dkeats.com/usrfiles/users/1563080430/repositories/wPres5.jpg][/ALERT]
[ALERT: url=http://www.dkeats.com/usrfiles/users/1563080430/repositories/wPres6.jpg][/ALERT]

Get Chisimba from http://avoir.uwc.ac.za

---

Institutional repositories: closing the chasm of apathy Tweet
608 days ago

Institutional repositories are interesting concepts. They are founded on the very principles that drive science, yet institutions struggle to implement them, and struggle to get researchers to contribute to them. Furthermore, most of the technology used for them is, like scientific publication itself, stuck in an antiquated 20th Century mode. This year should be the year that institutional repositories really take off in South African higher education institutions. But it probably will not, not for technical reasons, but for the same reasons that technology projects often fail: people and process reasons.

http://en.wikipedia.org/wiki/Institutional_Repository.

One of the things about Web 2.0 is the evolution of the notion of 'content through process' rather than 'content after process'. Institutional repositories are stuck in the 'content after process' mode, and unless this changes, institutional repositories will generally remain a nice ideal that very few people live up to in reality.

Here are some suggestions for getting institutional repositories to actually work and function in an institutional context. They may not be popular suggestions with all researchers, who resist anything that might be considered 'managerialist' in approach, but there is no viable alternative other than to abandon repositories altogether and return to the delicious independent obscurity of the previous century.

So here are the interventions that I sugggest under the 'content through process' umbrella.

1. Research administration processes need to include an automated submission of research articles to the repository at the preprint stage. One way to accomplish this would be to request that preprints be included with funding applications for local research grants, conference travel fund or promotion applications. Software could easily be written (a couple of days work) to automatically add the document to the repository. Of course, hidden in this is the complexity of such a request, and the carrot against which compliance is effected.

2. Parse annual research reports and extract lists of articles that are not in the repository, and the author is emailed a request for a copy of the article for adding to the repository. The article can be added by attaching it to a return email, obviating the need for the researcher to do anything beyond a normal email procedure.

3. Search RSS feeds of published articles involving staff at a given instution are used to automatically add an article to a local database,  and follow the above procedure.

4. The DropBox method. Ask academics to install a dropbox tool so that they can copy files into a directory on their local computer and have them automatically assembled into a preprint and loaded into the repository.

5. More radical and managerialist, just like promotion would not happen without a list of publications, ensure that promotions processes also include review of contributions to institutional repositories.

6. Social marketing is one method of changing a culture. Institutions need to engage in activities that change the dominant institutional culture from the currently rampant 20th Century version to a more 21st Centure, open, collaborative one where these activities are done because they add value.

Unfortunately, instutional repositories only become valuable when large numbers of people are using them. Then their value is obvious, as physicists for example have discovered. The trick has to be to engineer a tipping point within individual institutions, and that is going to require a variety of tricks, only one of which is the provision of software tools. I will make another post in a day or two in which I will talk about some of the technology innovations that are still necessary for Repository 2.0, and a further post on the humble presentation and makinng a repository for it. Meanwhile, here is an interesting presentation to look at:

 

I called this post 'closing the chasm of apathy' becuase apathy arises out of not seeing the potential of engagement, or because the effort of engagement is higher than the perceived benefit, or out of general ignorance of the object of unconcern. We must use better technology, smarter processes,  a little bit of managerialsm and social marketing to close that gap. At least in South Africa.

For the oh duh! people, there are many things that I have not said in this article. That does not mean that I do not think they are important. For example, it is obvious that the 21st Century repository must have semantic elements -- thats why I included the presentation above. I will talk more about that later.

---

If wishes were horses, data centres would run Linux clusters Tweet
609 days ago

The article by Lev Gonick in Inside Higher Ed ( http://www.insidehighered.com/views/2009/01/06/gonick ) catalysed my frist steps back into serious blogging for 2009. In my previous blog post, I mentioned the 11th prediction that Gonick made, and said I would talk about it later.

The 11th prediction was that the campus data center goes under the scope. Specifically, he wrote:

In South Africa, data centres are also under pressure, and the audit regime is increasing the costs of looking after them from the perspective of disaster recovery and business continuity planning. We have been raked over the coals a number of times by the audit process because of inadequate DR, even though our revenue is inadequate to cover the costs of full DR and business continuity processes. The only way to have adequate DR would be to shut down certain operations that are vital to the institution, something that the audit process fails miserably to consider. I suspect this is the case with most South African universities.

The power consumption of data centres should be under scrutiny in South Africa, given that fact that Eskom (our electricity monopoly) are unable to generate enough power to meet national consumpion. We should be looking at data centre optimization, but we are not, and we are unlikely to do so in 2009 to any meaningful degree. The biggest threat to green data centres in Sough African higher education is the over-reliance on stand- alone servers. This is in part because of the proprietary software technologies in use, and the high costs switching to more efficient software, and partly because of the mindset of the IT community in South African HEIs.

Since most servers are standing idle or using less than the optimum CPU cycles most of the time, there are tremendouse opportunities for improvements in efficiency. To make our institutions greener, we would expect to see a decline in the use of stand-alone servers, an increase in the use of blade technologies, and especially a tremendous increase in clustering and virtualization. We have begun to do some of this at UWC in some areas of operation. We have doe away with nearly all of our stand-alone servers, and have gone for blade technologies. We have also started to cluster and vvirtualize those applications that lend themselves to virtualization or clustering, something that I assume continue in 2009. But the kinds of software applications used to run the administrative support tend not to lend themselves well to clustering, so continue to run in blades as stand-alone blades. But at least there is some efficiency improvements happening.

The Hardware as a Service (HaaS) and Software as a Service (SaaS) opportunities mentioned in my last blog also provide for improved data centre efficiencies. However, in 2009, there will still be bandwidth impediments to implementing them. However, they need not all involve third party services. Since we are all facing the challenge of DR and business continuity, imagine the value, for example, of a central service for DR that might be hosted on a National Researche and Education Network (NREN) for example by TENET - the Internet Service provider owned by HEIs in South Africa. What other HaaS and SaaS services might we be able to leverage as a national system if we put our collective heads together to plan for them?

Those institutions which grab this bull by the horns, and embark on these kinds of efficiency improvements will have a leading edge over those which do not do so. This leading edge will have financially measurable value over the next few years, and others will struggle to catch up.

---

The Year ahead in Higher Education IT in South Africa Tweet
610 days ago

Lev Gonick, Vice president and CIO of Case Western Reserve University in Cleveland, Ohio, USA, wrote a piece in Inside Higher Ed, which appeared to day and gave the top 10 IT trends in higher education for 2009 (meaning, in the USA). The article sets out these trends against the economic downturn. The article is at http://www.insidehighered.com/views/2009/01/06/gonick .
 
The 10 trends are as follows, re-worded by me so that they make sense out of context of the article.
 
1. Major increases in the use of Internet-based, cloud computing for a variety of services
2. A deluge of consumer technologies penetrate the campus
3. Streaming media goes mainstream
4. SecondLife becomes a significant collaborative platform with new pedagogies to support it
5. e-book readers distrupt the college textbook market place
6. The IT helpdesk becomes an enterprise service desk
7. Most institutions move to open source 'course management systems'
8. Campuses hold back on ERP upgrades, and look at shared hosting of ERP systems
9. Requirements for better decision making will bring 'business intelligence' to the fore, including both tools and metrics-based decision making processes
10. Interactive, high definition videoconferencing moves to the lecture hall.
 
He also adds an eleventh one, that datacentres will go under the scope from a variety of perspectives, including budgets, energy consumption and space use.
 
In a few weeks, I will be moving to the University of the Witwatersrand, in Johannesburg, to take up a Deputy Vice Chancellor position for Knowledge and Information Management. It would be interesting to look at how these proposed trends might play out in South Africa.
 
1. Major increases in the use of Internet-based, cloud computing for a variety of services.
 
Cloud computing and other Hardware (HaaS) & Software as a service (SaaS)  opportunities are only beginning to make an inroad into higher education institutions in South Africa. At UWC, we are in the process of switching our student mail to the Google offering, have outsourced our spam filtering to two different software services (one for incoming, another for outgoing). UNISA (the University of South Africa) are also making a similar move with a service from Microsoft for to student email.
 
This trend will certainly continue in 2009, but such services require a 'decent' (sic) amount of bandwidth, and in South Africa, thanks to ill-considered national policies and ultraslow changes, we only have indecently expensive and therefore limited bandwidth. It should begin to improve mid 2009, but it will not happen fast enough for HaaS and SaaS offerings to make major inroads in South African HEIs in 2009.
 
However, those institutions which are able to identify key opportunities in this space will be ahead of the pack and will be able to continue their lead in 2010.
 
 
2. A deluge of consumer technologies penetrate the campus.
 
In South Africa, the economic reality of the majority of students in HEIs is such that consumer technologies are less pervasive than they are in the USA. However, cellphones are pervasive, as are technologies such as mXit that enable messaging over GPRS at a cost much lower than SMS technologies. At UWC, we have deployed these technologies in our e-Learning platform, and have participated in a project to use them for Drug Information and counseling. Over the Christmas period, over 50000 messages were exchanged on the system, thus showing its ubiquity and potential. In addition, a lot of cell phones have MP3 play capability, or students have simple, cheap MP3 players. At UWC, we have been building classroom-based podcasting systems using cheap devices and Free Software (Open Source) applications, some of which we have built ourselves.
 
Fairly basic cellphones and MP3 players are the consumer technologies that have the most potential in HEIs in South Africa. Those institutions that are best able to exploit messaging and MP3 audio capability will have an edge on this space because they will have built the educational processes that can be adapted to new consumer technologies as they become available.
 
3. Streaming media goes mainstream
 
Again we are limited by being in a bandwidth-constrained backwater on the Internet. Streaming media technologies will begin to make inroads, using services such as YouTube and others, especially now that caching of such media is feasible. However, real streaming is still going to be limited by bandwidth in 2009.
 
Those institutions that begin to experiment with video technologies, and streaming in 2009 will have a lead in 2010 when bandwidth is less of a limitation. This will again be mainly because they will have developed the pedagogical and educational processes that can be adapted as the bandwidth environment improves.
 
4. SecondLife becomes a significant collaborative platform with new pedagogies to support it .
 
Use of such high bandwidth requiring technologies in South African HEIs will remain experimental in 2009. This is not only due to bandwidth requirements, but because producing content for such immersive environments is costly, and difficult. Other priorities will dominate 2009 IT expenditure.
 
5. e-book readers distrupt the college textbook market place.
 
At UWC we have been experimenting with e-books as part of the IADP project, and it has been fraught with problems, mainly due to the so-called 'digital rights management'. Given that e-book readers are not widespread in South Africa, and the highly restricted nature and high cost of e-books, this disruption is unlikely to happen in South Africa this year. The widespread practice among students of ignoring copyright and photocopying everything is likely to continue. This might change in 2010, but only if the publishers are able to come up with a cost model that takes the non-rivalrous nature of digital versions into greater consideration.
 
6. The IT helpdesk becomes an enterprise service desk
 
Many universities in South Africa are struggling to come to grips with having a working IT service desk. While I have proposed this at UWC, the institution is not ready for such a major leap quite yet. I expect that other South African universities are in a similar boat. Instead, 2009 will be the year in which we begin to share much more of our service desk experience, and learn from one another to the point where we mature enough as institutions that we can begin to look enterprise wide in 2010 or 2011.
 
7. Most institutions move to open source 'course management systems'. 
 
This has already begun to happen. Universities in South Africa use Sakai, Moodle and KEWL, but proprietary systems such as Blackboard and WebCt are still common.  At UWC we use KEWL, developed as a e-learning platform, not a course management system, with Web 2.0, Education 3.0, cloud computing and distributed applications in mind. For example, KEWL 3.0 can run in the Amazon Cloud, enabling institutions to immediately leverage the benefit of cloud computing and hardware as a service. It can also make use of distributed functionality form over 30 Web 2.0 services and applications. This trend will happen in other systems as well, but the architecture of some will slow down their ability to adapt. 
 
Those institutions which are best able to leverage Free Software (open source) in combination with distributed services will have an edge in 2009 because few will achieve it.
 
8. Campuses hold back on ERP upgrades, and look at shared hosting of ERP systems
 
There is no consistency among campuses in South Africa at present with respect to the ERP systems, especially those used to manage students. At UWC, I had tried and failed to develop a Free Software student ERP system, not because of the technology, but mainly because of mindsets that could not be changed.
 
HEIs in South Africa will continue to invest in the proprietary systems that they are locked into in 2009, and there will be very little change in this space. However, a conversation should begin among the executives (DVC, Executive Director) responsible for IT in HEIs in South Africa towards the creation of a shared, Free Software (open source) ERP for HEIs. This is not only consistent with the national open source policy and strategy, it makes such uncommonly good sense. However, the HEI system is perhaps not ready for this level of collaboration. Otherwise, in SA, it will be business as usual with respect to ERP systems, with attempts being made to reduce costs wherever possible. Vendor lock-in in is strong, and the vendors will attempt to exploit that lockin to ensure that they retain their customers. Many of them will be very aggressive about it.
 
9. Requirements for better decision making will bring 'business intelligence' to the fore, including both tools and metrics-based decision making processes .
 
The national higher education management information system (HEMIS), and the increasing need for efficiency, accountability and effectiveness, means that this trend will affect HEIs in South Africa in a big way. Those that do not take this seriously will fall behind in their ability to optimize their government funding, and will be taken to task by their councils and auditors. Increasing demands for just-in-time information being available where it is needed, when it is needed, will spur improvements to both reporting tools and the processes whereby the information is used.
 
10. Interactive, high definition videoconferencing moves to the lecture hall .
 
This will not be a significant trend in South Africa in 2009. The technical hurdles will be too great, including the high cost of equipment, the high cost of production, and the bandwidth limitations. Selected universities that have invested in analogue television to aid distance learning will continue to invest in the video area, but it will not be a major trend across all HEIs. 
 
Instead, most HEIs will continue to invest in improving the basic technologies for use in the digitally enhanced classroom, including basic computer, Internet and projection facilities.
 
With respect trend 11 regarding data centres coming under scrutiny, I will leave my comments on that for another post. There is a lot to say there....

---

Experimenting with night photography Tweet
613 days ago

During the holidays, I have been experimenting with night photography, which has been quite a lot of fun. I use a Canon EOS 50D camera, a 30 year-old broken Manfrotto tripod, and a remote trigger for releasing the shutter.

---

Amarok no more: Ubuntu Intrepid woes Tweet
613 days ago

I rue the day I upgraded my desktop to Ubuntu Intrepid. I had a perfectly working Hardy installation, so why oh why did I want to fix something that was not broken......

I previously wrote about how I was enjoying the Amarok nightly updates via their Neon service. Sound on Intrepid has been an absolute pain, and I finally have everything working after hours of labouring over stuff that was never a problem before. But I cannot get sound out of Amarok, neither the nightly  builds nor the stable release. I have wasted hours on it, read every discussion post, IRC log, and webpage I could find, but nothing works. A mute music player is of little use, so I have removed it for the time being, and gone back to a less perfect music player, but one with sound!

---

Testing the Flock Blog editor Tweet
616 days ago

Having just reinstalled Flock, I thought I should play around with some of its features. One of them is a blog editor, so I am trying it out here with the CHisimba implementation of the Metaweb API. This post started out being just the first two paragraphs, but 'stuff' intervened.

I like the editor because it is a tool icon in Flock, and it gives me a nice fat, full-screen, desktop-like editing experience, at lease when it does not crash. It also allows source inserts, which is vital for inserting widgets and snippets (like the Get Flock) button below.



One of the problems with using a browser based blog editor, however, is that it is dependent on the browser. And new browsers are nortorious for crashing. In my case, I typed all of this once, and was busy editing while uploading some pictures to Flickr. The Flickr uploader is about as stable as an insane person on a bad day, so it naturally crashed the browser, and brought down my editing session.

I was left with no option but to open a text editor and retype what I had already typed, by reading the greyed out text from the crashed screen. This is of course one of the annoying 'features' of browsers that Google is trying to fix with its Chrome browser, the tendency of one crashed site to bring down the entire browser and all other sites in other tabs. Hopefully, they will eventually succeed, and other browsers can do likewise.

When I restarted Flock, I was presented with a dialogue asking if I wanted to recover the post that I was typing when it crashed. I did this, but the resulting content was not editable, indeed, Flock appeared to have crashed despite the fact that I had chosen a new session. I closed Flock, and tried to reopen it, but it said it was already running.

Running ps -ax | grep flock showed a whole bunch of running processes, but no Flock on my desktop. I killed all four Flock processes, and this time Flock opened again.  It still gave me the option to recover the dead file. Should I or shouldn't I? I decided to let it recover, and this time it worked. So, here I am testing it at last!

Yes, it worked but I had to go to the site to edit it because:

  1. It does not have a Creative Commons chooser.
  2. It doesn't insert the tags into the tags field in the blog.

So, for me, it is not very useful. It also failed to find any categories, but I suspect that might be a bug on our side as none of the blog posters I have tried found categories either.

Tags: flock, linux, blog

---

Use installed Firefox plugins for Flock Tweet
616 days ago

I just updated my computer at home to Ubuntu Intrepid, and reinstalled the Flock social browser on it. Since I already have Firefox installed, there is no need to install the plugins again, just link them to the Flock plugins directory.

On Ubuntu Intrepid, as at Jan 1, 2008, Flock plugins seem to be in /usr/share/flock/plugins according to grep results:

   desktop:~$ locate flock | grep plugins
   /usr/share/flock/plugins

Therefore

   sudo ln -s /usr/lib/mozilla/plugins/* /usr/share/flock/plugins/

is what worked for me on Intrepid. Thought I may as well share that here since it may be useful to other Flock and Firefox users. Of course, if your plugins are in Flock first, you can also link them to Firefox using the reverse of this command.

Note: this works for pluginss. It is not suitable for extensions!

---