Another semester gone …

Wow time flies indeed. It is June already and I am thinking what has happened to the year? I wish I knew an answer to that one. I don’t have much to show except perhaps a stronger sense of where I am going. Until yesterday, I wasn’t really sure about my destination but now the picture is clear. And this is thanks of course to great supervision 🙂 .

Over the past months I have had the privilege of understanding why people claim that age brings wisdom. Alf has repeatedly told me that sometimes hating your work is a signal of working… and naturally I thought that was insane. I mean how could going nowhere quickly be regarded as working and most importantly part of the journey? I didn’t understand that before but today all the little pieces are coming together! And as Alf has once said: they could be better ways to cook pasta than methods known to us presently so continuing the exploration is not an exercise in futility. With that I continue with my journey and thankfully with a stronger appreciation of taking wisdom from the old 😉 .

Visit to Raphael Centre

Paid a visit to the Raphael centre with Denis. We had a chat with Sister Matama. After providing a brief overview of the system that we want to develop, we asked the type of questions that are commonly asked by clients. We were told that the following were very typical:

  • Will the test be painful i.e. will it be sore? [Answer: one gets pricked so it MIGHT be painful]
  • How long will it take to have the test and get the results? [Answer: 5 – 20 minutes but people need to be aware there is also the pre and post counselling that is mandantory that may affect the time]
  • Is confidentiality assured? [Yes it is. …]

For people that call to make enquiries, the most asked question is about the whereabouts of Raphael Centre and the times that it is open. Sister Matama emphasised that rarely do they get people who want to be counselled over the phone. That is, people seem to prefer to come in and ask about testing and other related information.

On the question of what our system needs to capture, a question asking the user about knowledge of their status should be the backbone of the system functionality. The answer to this question should:

  • provide information about places for testing.
  • put emphasis on confidentiality.
  • inform the caller that couples can be tested together but friends cannot.

Interview at Ragland Clinic

Just went for an interview at the Ragland clinic. After briefly providing an overview of what I would like to achieve, the following were deemed as important questions:

  • Where is the service provider located at?
  • What services are offered by the service provider?
  • What time are the services offered at?
  • What is next after I get HIV test result be it is positive or negative?

A question that might be asked to the caller is: what made them want to test? The cited possible reasons include:

  • Been encouraged to test by a health professional
  • Partner tested positive
  • Check of status (This is more a likely a response from the youth. The old tend to test on account of partner testing positive or being sick and have been encouraged to test.)

Contacts: Noel Issacs, Mandisa Nduna, Lungi Masingili, Pumeza and Ntozi

Advice from SAICIST Postgraduate Symposium

Today, we presented our work-in-progress papers at SAICIST Masters and Doctoral Symposium. It was truly very interesting and these were some of the general comments that were made by the moderators:

  • Articulation of the problem statement: It was felt that many of us failed to state our problem statement.
    Research question formulation: Again, we were faulted for articulation but the major criticism was with the formulation of the question. The formulation of the question tended to be binary as in yes or no. The problem with this is that contribution to the body of knowledge is not highlighted as a priority.
  • Motivating the relevance and importance of the research: We were reminded that the responsibility lied with us to sell our research. So it was important to address the WHY at length.
  • Claiming one’s contribution with panache: It was recognised that many people showed reluctance in claiming their contribution. However, it was cautioned that when they do they should do it with humility and perhaps a touch of panache.
  • Communicating stage of research: In presenting we failed to inform the audience the stage we are in of the research. That is, are we at the start, middle or end? Consequently, the audience could not gauge what types of questions to ask.
  • Confidence: While it was understandable that many of us lack experience in presenting but we were urged to show confidence. After all, it is our research and odds are we know more about it than many people in the audience.
  • Differentiating software product from research: This was directed specifically to the computer scientists. Apparently we tend to think that the product that we will develop is the ultimate research. We need to understand that whatever we develop is but a by-product of research.

All in all, the important message to take home was explicitly addressing the WHY, WHAT, HOW and last but no least the SO what questions in the presentation. I must say it all sounds cliché but I guess one can never overestimate the old wisdom of sticking to the basics!

Asterisk and audio formats

Gosh I sure know how to derail myself! Today, I spent a good chunk of my time investigating what is the best format and codec to use for my audio files, which are part of the voiceXML application. This was not meant to be a time-consuming exercise since Asterisk technically supports many formats and codecs. However, I was keen on finding something with a low bandwidth footprint and reasonable quality. To be precise I was thinking of something along the lines of 8 bits sampling size and sampling rate of 8 KHz.

Why? Well because I am bandwidth sensitive but also because ordinary voice doesn’t necessarily benefit much from high sample rates (or at least this is what you get if you follow debates related to podcasting for developing worlds). Ok back to the point 😉 ! In my adventure I came across speex and I must say, I am so in love with it. It truly seems to be something that one must consider just check out how it compares with other codecs by following this link What is also very interesting is the modesty attached to the comparison as captured by a disclaimer which basically says the results should be taken with a grain of salt!

The only issue I have at the moment is that speex is not supported by audacity. Thus after file creation one has to encode the files to .ogg files at the command line. This all translates to not being able to kill two birds with one stone which is not good because I certainly believe in killing two birds with one stone (just as a side issue I have been provided this link by Richard as part of dissuading me into such a believe). But when it comes down to it, I suppose my issue is not even an issue because at one point or the other one might need to convert files for Asterisk using command line utilities like SoX, “the Swiss Army knife of sound processing programs”. So when all has been said and done, I think I shall be a speex fan and not fanatic 😉

Departmental Seminar

I was presenting today in the departmental seminar. I was rather anxious to get it over and done with. So I “persuaded” Mos to allow me to go first. I think I didn’t do that badly except that I was fast and at times unable to articulate myself 🙁 . Consequently, I failed to explain the architecure of the service. Further, I didn’t really provide enough details on the types of questions that can be answered by the ontology underconstructions. That is, I failed to communicate what will constitute the competency questions of the ontology, thus the type of queries it would handle.

On the positive side I can take today’s seminar as a dress rehearsal for SATNAC. The question then that remains is whether with unfamiliar faces I would be able to slow down a bit and articulate myself better. That remains to be seen but I am an optimist 🙂

VXI* VoiceXML browser

I have recently started playing with VXI* VoiceXML browser. Granted that it is a commercial product and we might not afford to buy a license, I thought I should give it a try. One of the reason is that it is designed specifically for use with Asterisk.

Installing it was the easy part but getting it to work was another story. For reasons unknown to me, using the start up scripts to start and stop the browser was a hassle. Specifically, starting the browser with the command /etc/init.d/openvxi stop didn’t work. After several exchanges with the tech support team at I6NET Solutions and Technologies we found a workaround or rather we resorted to starting the browser manually using: /usr/sbin/openvxi -channels 100 -config /etc/openvxi/client.cfg &

I am now happily playing with the browser although I wish the documentation could have included the option to start the browser manually. All is well for now 🙂

Unintended ‘Imperialist’

As part of building the ontology for capturing knowledge about HIV and AIDS, I came across UNAIDS’ Terminology Guidelines document. The document presents preferred terminology for discussing matters on HIV and AIDS. For example, response to AIDS is the prefered alternative to fight against AIDS.

There are many examples, but one in particular, made me realise that in writing my proposal, I came across as an imperialist of sort! I used the term intervention a number of times in articulating the goals of my work. As aptly put in the document,

“[t]his term conveys “doing something to someone or something” and as such undermines the concept of participatory responses.”

In some circles, this translates to imperialism. So in such circles all I can really say in my defense is: I was but an intended imperialist in writing the proposal. I will in the future try as much as I can to avoid terminology that may be perceived in any negative light 🙁