Fortnightly Mailing

Machine translation reaching a tipping point? An example of the improvement in the quality of Google Translation.

I happened to be looking at a piece of rather vernacular one-finger-typed Spanish that I had translated into English with Google Translate in 2005 and 2007. There'd been no change in the translation between 2005 and 2007, and the results [A] were of partial value, with some entertaining gibberish.

Today's translation [B] is very much better, remarkably so, as you will see below. There is a translation of the same piece by a native speaker of Spanish is included at C.

Original Spanish

1. Una vez al año se hacia una reunion en algun balneario o algo por el estilo con yodas las familias.

2. Otro punto era que se insistia que la reunion era a (calzon quitao) quire decir que tenes que poner todas las cartas sobre la mesa (no esconder nada) muchas veces requeriamos la presencia de la señora para ver que opinion tenia ella.

3. Los cabezas de grupo al final de la reunion exponian todo lo que su grupo recomendaba y a veces unas verdades o criticas que dolian mucho y era esto que me parece que ayudo a muchos a ver la realidad.

A. 2005 and 23/5/2007 translations by Google

1A. Once a year towards a meeting in some bath or something of the sort with yodas the families.

2A. Another point was that insistia that the meeting was a (trousers quitao) quire to say that tenes that to put all letters on the table (not to hide anything) often requeriamos the presence of the lady to see that opinion tapeworm she.

3A. The heads of group at the end of the meeting exponian everything what its group recommended and truths sometimes or you criticize that they dolian much and was this that seems that I help many to see the reality.

B. 28/7/2010 translation by Google

1B. Once a year to a reunion at some spa or something like that with yodas families.

2B. Another point was that he insisted that the meeting was to (take off your breeches) quire that you have to put all the cards on the table (not hiding anything) often required the presence of the lady to see that opinion had it.

3B. The group heads at the end of the meeting setting out all that his group recommended and sometimes some truths or criticisms that hurt a lot and it was this which I think helped many to see reality.

C. Human translation

1C. Once a year there was a meeting at a swimming spot or place of that kind, with all the families.

2C. Another point was the insistence that the meeting should be “with underpants removed” which means that you have to put all your cards on the table and hide nothing, and often we required the presence of the wife to sound out her views.

3C. At the end of the meeting the leaders of the group reported on all their group recommendations and sometimes there emerged some truths or criticisms which were very painful, and this is what I think helped many to come to terms with reality.

Assessing the Effects of ICT in Education - an OECD report

Assessing the Effects of ICT in Education - Indicators, Criteria and Benchmarks for International Comparisons is a 200 page report from the OECD's Centre for Educational Research and Innovation. It is available for free download from the OECD Online Bookshop.

Abstract:

"Despite the fact that education systems have been heavily investing in technology since the early 1980s, international indicators on technology uptake and use in education are missing. This book aims to provide a basis for the design of frameworks, the identification of indicators and existing data sources, as well as gaps in areas needing further research. The contributions stem from an international expert meeting in April 2009 organised by the Centre for Research on Lifelong Learning, in co-operation with OECD (CERI), on benchmarking technology use and effects in education. The contributions clearly demonstrate the need to develop a consensus around approaches, indicators and methodologies. The book is organised around four blocks: contexts of ICT impact assessment in education, state-of-the-art ICT impact assessment, conceptual frameworks and case studies."

Putting CIPD straight

Earlier this month CIPD - a large membership organisation with a royal charter - published a shoddy report about spending on education and training quangos.

The report was shoddy because:

• several of the organisations identified - including LSN and NIACE - were not quangos at all;
• the figures cited were inaccurate;
• the author of the report seemed neither to have understood the work of the organisations in question, nor to have checked his facts.

CIPD subsequently withdrew the report, but not before The Daily Telegraph had blithely repeated the report's assertions.

Here are links to justifiably angry rebuttals from LSIS [100 kB PDF], NIACE, and LSN. Excerpts:

NIACE - "NIACE's reputation, built up over eighty nine years, rests on its independence. Unlike trade or professional associations, NIACE does not defend a single sectional interest and unlike a quango, it can and does campaign against government policy on occasion (for example by drawing attention to the loss of 30 per cent of publicly-funded places for adult learning between 2004 - 2008). This advocacy work is not funded from the public purse but through other charitable funds - which are also used to fund independent research, such as the Inquiry into the Future for Lifelong Learning that reported last year."

LSIS - "It is unfortunate that an organisation such as the CIPD that purports to support professional development can be both so inaccurate and dismissive of this activity. It’s a pity too that they can be so unprofessional as to not check their facts or find out how an organisation really works or what it does before rushing into print. I’m sure many of their existing members will be very disappointed in what is clearly a lack of professional competence."

LSN - "LSN is not a quango.  It is a private, independent charity which seeks to improve learning and skills in this country.  Contrary to the CIPD’s report, LSN is not in receipt of grant funding, nor have any of its recent acquisitions been funded by the taxpayer."

Teaching and learning tool to create mobile applications for Android-powered devices

I covered App Inventor for Android in Fortnightly Mailing last August soon after Hal Abelson, on secondment from MIT to Google, announced it.

App Inventor:

"is a new tool in Google Labs that makes it easy for anyone—programmers and non-programmers, professionals and students—to create mobile applications for Android-powered devices."

App Inventor is now publicly available as a "beta", and anyone can now request an App Inventor account. App Inventor is primarily designed to support teaching and learning computer science, in the tradition of the Logo programming language, which started life over 40 years ago.

Instinct or Reason: How education policy is made and how we might make it better

CfBT's Instinct or Reason: How education policy is made and how we might make it better by Adrian Perry, Christian Amadeo, Mick Fletcher, and Elizabeth Walker "investigates the factors that lie behind the formation of educational policy". It is "based on discussions with an expert group, a desk based literature review (including academic research and politicians' memoirs), interviews with stakeholders and an extended process of draft revision".

You can download the full report from the CfBT web site [1 MB PDF], and there is also a 23 slide presentation [0.5 MB PPT] from the report's launch on 7 June.

Meanwhile the report's main recommendations, which make me think that in less straightened times there'd be mileage in the establishment of an Educational Research Council along the lines of the Medical Research Council, are as follows.

(a) The recommendation that the prime role of ministers is to bring their values to inform goals and ambitions, rather than tactics and methods, where expert analysis should play the larger role.

(b) An expert commission, analogous to NICE in healthcare, should be established to create and interpret educational research, evidence and analysis. Such a body should advise institutional leaders as well as politicians and civil servants. Ministers would be encouraged to share their thinking when their analysis differs from that of the commission.

(c) An office of Chief Officer – analogous to the Chief Scientific or Medical Officer – should be established. He or she should build strong links with the Select Committee system.

(d) Evaluations should be independent, commissioned outside the Department and published. Research and evaluation should be brought together to share a budget.

(e) Given the short career life of ministers and the limited life of governments on one hand, and the need for long-term implementation of educational reform on the other, there should be a search for consensus between political parties on non-controversial ground.

(f) Attention should be given to the perception that little useful research is being generated for education policy makers. We recommend that a portion of the budget for educational research should be directed to topics which can be seen to relate closely to identified needs of the system.

(g) Researchers should remain independent, but be given help to present their conclusions in a way that will give the best chance of calm consideration rather than rejection. (h) A prize should be established for well evidenced policy.

(i) Better links should be built between practitioners, researchers, civil servants, politicians and quangos – represented in shared career paths.

(j) International comparisons should be encouraged as part of a managed learning system.

What research has to say for practice - nine guides from ALT

ALT (for which I work half time) has made public nine What research has to say for practice guides. The guides, which are freely available, and which are open for others to edit, have been written mainly by members of the ALT Research Committee. It remains to be seen whether the guides languish without contributions being made to them, or whether people in the learning technology community decide to "muck in" and maintain and develop them.

Becta's Emerging Technologies web site, and a cross-site search

Becta currently runs a big Emerging Technologies web site. It has valuable contents but I found it frustratingly organised. (I vaguely remember being consulted a few years ago - with wire-frames - by a company that had won the contract for this service, or the web site on which the service runs.) I found the site quite difficult to navigate, with unhelpfully short abstracts, and peculiar policies on things like the dating of items, and their authorship. I n the continuation post below I combine some links on the site with abstracts about the (interesting and useful) items to which they point.

Abstracts of this kind save the user a lot of time in the long run, and JISC does them a bit better in sites like this one, although it is also evidence that the UK technology in learning world is unhelpfully fragmented, something that is on my mind as a result of UK Government's decision to close Becta.  (The Association for Learning Technology - ALT - for which I work part-time, had  something about this published yesterday.)

I played about with a custom search that gives returns from JISC, Becta, ALT, Naace, Futurelab, TEL  (it would be simple to add others, cheap to get the advertisements removed, and the current version of the code for the search window is at the foot of the continuation post if you would like to reuse it):

Loading

10 techniques to increase retention (in memory rather than on courses)

Donald Clark has posted a terse and informative 10 techniques to increase retention, which deserves to be widely read, though (if this is does not seem churlish) it could do with a couple of links to references for each of the techniques.

Teaching with games - Katie Salen from New York's "Quest to Learn" school explains

Below is a gripping 7 minute video in which Katie Salen, design director of the Quest to Learn school in New York. Full of choice quotes about why game design is such a powerful medium for learning, for example: "To design a game you have to really know what you are talking about in order to create a system that models that idea."

The video is one of many on the Intel-sponsored Big Think, amongst which is this talk by Leonard Kleinrock about the invention of the internet protocol. This struck a very strong cord with me having as a child gone through the make-your-own-crystal-set-from-bits-around-the-house stage, including the visit to an electronics shop to buy a variable capacitor, as described by Kleinrock. There is a connection between the two videos in that people of my generation had many opportunities to tinker - in the 1950s to 1970s, and to learn a lot of science and maths from this. Today the scope to tinker with technical objects is more limited because they tend to be much more "sealed".  Is a games based curriculum a modern equivalent of the home made tinkering curriculum of old?

Links to articles in the May 2010 ALT News Online

Here are links to the main articles in the May issue of ALT News Online, edited by Morag Munro:

• OER 2010: Exploring OER Content, Design and Communities, by Theo Lynn and Neil Bruton;
  
• Blackboard Teaching and Learning Conference 2010, by Leo Havemann and Mimi Weiss Johnson;
  
• Telling Stories at ELI2010: the Educause Learning Initiative 2010 conference, Texas, by Jackie Carter;
  
• MoodleMoot UK 2010, by Jessica Gramp;
  
• MEC 2010: Celebrating 30 Years of Sharing Learning Technology Experience and Expertise, by Theo Lynn, Neil Bruton and Kieran Linehan;
  
• Innovative Teaching and Learning Research - Studying Changing Teaching Practices and Students' 21C Competencies, by Maria Langworthy;
• Confederation of Open Access Repositories Inaugural General Assembly, March 2010, by Neil Jacobs of JISC;
  
• Using a departmental blog with post-16 Health Sciences students, by Keith Shaw;
• Online Learning – a Guide for Associations, by Louella Morton and Brendan Noud of WBT Systems;
  
• Action Research as a key to stimulating Innovation and Professional Development in ILT, by John Webber;
• Amplified staff development, by Jennifer Jones, Joanne Badge, Stuart Johnson and Alan Cann.