Semantic Werks

We camped (tent camping, aka ‘real’ camping, albeit car supported) in France (SW, central, and NW) in late September 2007. Here are a few lessons learned.

First, brush up on driving stick. I had a few problems early on that resulted in a smell of burned transmission fluid lingering for a few days. At the same time a ‘how to drive roundabouts’ refresher might be useful too. A useful post I read after the trip mentioned you should factor in paying the same in tolls as you do in gas, which seems about right. We had a diesel Opel, and a litre of diesel was approximately 1.20E.

The time of year was pretty great. We encountered patchy periods of rain and sun, but no lengthy rainy periods. Most flowers were finished, so we drove past fields of zombie-like dead sunflowers and bare fields, but they had a certain charm.

In general, we did too much driving: from Paris, to Vimy Ridge, to Normandy, to Bordeaux, to Lascaux/Vezere, to the Loire, and back to Paris. Next time, I would consider taking a train to a major hub, such as Nice, and then renting a car for a few days to explore. It’s nice to return to a pitched tent at the end of the day.

Campsites in France are plentiful and cheap. Sites are generally a grassy parking spot, as seen above, and I can imagine at peak times the campgrounds could get quite noisy. At this time of year, while a few were closed, the campgrounds were largely empty and quiet. Costs are from 10-12 E for two people with a car. We basically winged it, just picking towns with a certain size (based on the font size on our map!) and looking for camping signs. We were only burned this way once, and that’s because the town was pretty tiny. Luckily they hadn’t locked the entrance, so we pitched the tent on the grass anyway. Here are the places we stayed:

• Municipal camping in Cambrai, west of the main river through town. Generally the municipal campgrounds seemed to be near industrial parks where land is cheap. Non-municipal sites are more like resorts for vacationing Europeans with campers. HQ for visit to Vimy Ridge.
• Municipal, closed camping in Tremblay. After visiting Mt St Michel/Normandy.
• Municipal site in Mirambeau, for three nights. Very central for Bordeaux, Cognac, Blaye, etc.
• Municipal site in Uzerche. Very pretty site on a river.
• Municipal site in Amboise. We only spent a short time here, but it would be a great HQ for exploring the Loire. Beautiful, too, right across from a gorgeous chateau (below).

Cooking: MSR stoves, or any stove which burns white gas or naptha, is pretty much impossible to fuel in France (since bringing fuel on the airplane is a big no-no). All the Euros use the ‘Gaz’ brand stoves, and the canisters seemed relatively easy to find in larger shopping stores — ‘Super U’ or outdoors/garden centres. Next time I would just bring a cheap Gaz model from Canada. Generally I dislike the butane stoves, they don’t burn as nicely, and the containers are typically not recyclable.

Although I tracked down ‘naptha’ in a grocery store, it must have been diluted for cleaning only, as it didn’t burn, just evaporated. In the end I begged a gas station attendant to allow me to fill my fuel bottle from the pump with the unleaded gas the MSR Whisperlite can burn. This site also suggests looking for dry cleaning fluid (?). Unleaded gas is cheap but quite dirty. However, in six or so uses, I never had to clean the fuel line, so it was perfectly workable. You may run into trouble buying so little gas - I tried late at night at an empty station.

Camping was a great way to see a bit of the less-touristed parts of the country, and very affordable (allowing us to spend more on the meals!). Contact me if you would like more details.

I just returned from the ICSM conference and associated workshops in Paris, France (very nice, thank you). I have many notes on talks I saw, but herewith a few impressions:

ICSM is about software and machine artifacts, not requirements. Requirements come from on high, and the impetus for maintenance tasks is generally assumed to be well understood. It seemed a little like solving the problem of getting suburbanites to a downtown office by improving the highway signage, or improving offramps, but ignoring urban rail as an option altogether.

A discussion at the workshop on evolvability spurred me to consider the ’smart-monkey syndrome’. My contention is that building software, while hard, is not the hardest problem. I think there is a lot of evidence for this, starting with Brooks’s law, which is all about the external effects on software development, specifically communication.

Along the same lines, Bruce Eckel posted a neat thought-experiment, asking, “If Microsoft, with all the money and smart people it has, can’t release cool applications more than once every few years, than who can?” Several responses disagree, suggesting Microsoft is actually quite innovative. However, one of the respondents posted an argument with which I wholeheartedly concur, namely that it is again the external factors — multi-national lawsuits, legacy support for code from the 80s, internal communication, device support, etc. — that is the real bottleneck. This, I think, supports my contention that the building of software — e.g., GMail — is the easy part. I am almost certain that Microsoft engineers have had many of the same ideas as Google, but external reasons prevented their realization, such as corporate strategy, marketing, etc. Much as I have shied from the business side of software as fuzzy and pseudo-scientific (meaning that many of the claims in the literature are wholly unsubstantiated by evidence), I am coming to realize that it may be the one aspect that matters the most - and certainly more than language choice.

I came across a paper[1] via the Nebraska-Lincoln Empirical Software Engineering site that ranks software engineering schools and scholars. The authors assess importance based on citations in IEEE Transactions on Software Engineering, ACM Transactions on Software Engineering and Methodology, and the conferences ICSE, and FSE. I think most researchers would agree these are the top venues for software engineering, although I might quibble with the merits of the journals, since important new research is typically sent to conferences. Also, when a conference has an acceptance rate in the single digits, it tends to drive people to more specialized venues, which aren’t incorporated here, e.g., RE, ASE, Models, and so on.
The list is hidden in a PDF behind the ACM’s “Digital Library”, which isn’t a library at all, since users have to pay a lot of money to access it. So I’ve reproduced it here, because it’s interesting. One thing to note is that UofT, my school, is 26th, and behind UBC, Waterloo, and Carleton. There is a perception at UofT that we are not just the ‘best’ in Canada, but by far the best. Clearly, like Avis, we need to try harder. At least in these four venues.

Top 50 institutions

Rank Institution
1 Massachusetts Institute of Technology
2 Carnegie Mellon University
3 Georgia Institute of Technology
4 University of Maryland, College Park
5 Oregon State University
6 University of California, Irvine
7 University of British Columbia, Canada
8 Politecnico di Milano, Italy
9 University of Texas, Austin
10 IBM Thomas J. Watson Research Center
11 University of Waterloo, Canada
12 University of Massachusetts, Amherst
13 Imperial College London, UK
14 University College London, UK
15 Carleton University, Canada
16 University of Paderborn
17 Purdue University
18 Stanford University
19 Kansas State University
20 Katholieke Universiteit Leuven, Belgium
21 Michigan State University
22 University of Pittsburgh
23 University of Colorado, Boulder
24 University of Texas, Dallas
25 University of Washington, Seattle
26 University of Toronto, Canada
27 Ohio State University
28 University of Southern California
29 University of Karlsruhe, Germany
30 Osaka University, Japan
31 University of California, Davis
32 Fraunhofer-IESE, Germany
33 University of Virginia
34 Simula Research Lab, Norway
35 Washington University in St. Louis
36 Hong Kong Polytechnic University, China
37 Brown University
38 University of Illinois, Urbana-Champaign
39 University of Strathclyde, UK
40 NASA Ames Research Center
41 University of Bologna, Italy
42 University of California, San Diego
43 Avaya Labs Research
44 Northeastern University
45 West Virginia University
46 Case Western Reserve University
47 Rutgers University, New Brunswick/Piscataway
48 Bell Lab, Naperville
49 Institute for Information Technology at National Research Council, Canada
50 National University of Singapore, Singapore

Top 50 scholars

# Scholar
1 Harrold, M.
2 Rothermel, G.
3 Murphy, G.
4 Briand, L.
5 Ernst, M. (not related!)
6 Jackson, D.
7 Kramer, J.
8 Uchitel, S.
9 Mockus, A.
10 Egyed, A.
11 Magee, J.
12 van Lamsweerde, A.
13 El Emam, K.
14 Emmerich, W.
15 Chechik, M.
16 Batory, D.
17 Inverardi, P.
18 Devanbu, P.
19 Herbsleb, J.
20 Clarke, L.
21 Jorgensen, M.
22 Robillard, M.
23 Soffa, M.
24 Sullivan, K.
25 Letier, E.
26 Stirewalt, R.
27 van der Hoek, A.
28 Bertolino, A.
29 Dwyer, M.
30 Krishnamurthi, S.
31 Tonella, P.
32 Basili, V.
33 Kitchenham, B.
34 Taylor, R.
35 Memon, A.
36 Michail, A.
37 Dingel, J.
38 Notkin, D.
39 Walker, R.
40 Orso, A.
41 Roper, M.
42 Griswold, W.
43 Kemmerer, R.
44 Leveson, N.
45 Padberg, F.
46 Roman, G-C.
47 Sinha, S.
48 Tian, J.
49 Engler, D.
50 Elbaum, S.

[1] Jie Ren and Richard N. Taylor. Automatic and Versatile Publications Ranking for Research Institutions and Scholars. Communications of the ACM (CACM), Vol 50, No. 6 (June, 2007). http://doi.acm.org/10.1145/1247001.1247010

Update:

I came across another ranking of researchers. This one [2] is specific to requirements engineering. Here, the University of Toronto fares much better:

[2] Alan Davis and Ann Hickey and Oscar Dieste and Natalia Juristo and Ana Moreno. A Quantitative Assessment of Requirements Engineering Publications — 1963–2006. in proceedings of Requirements Engineering: Foundation for Software Quality, 129–143, 2007. [BibSonomy] URL