Mesquite 3.10 released


We’ve released version 3.10 of Mesquite, which has some important bug fixes and workarounds for operating system problems.  There are also some new features including a new alignment tool, the ability to BLAST a local database, a batch processor files in a folder, and the ability to read MrBayes sumt files.  For more details, see here.

MacClade’s 30th Anniversary

MacClade1Today is MacClade’s 30th birthday! Version 1 was released 21 June 1986, with David joining me for version 2 the following year.  The last release was version 4, but it’s not commonly in use today, as it is no longer compatible with modern operating systems. Those biologists who use it need to maintain ancient machines (or emulators) to run it. There are, however, some who do use it, and we often get comments that people miss it. Some of its DNA persists in Mesquite (both conceptually and as some code translated straight from Pascal into Java).  Happy Birthday, MacClade!

A proposal for pronoun gender for AlphaGo and other AIs

Commentary on the ongoing historic Go match between Google DeepMind’s AlphaGo program and Lee Sedol has included some anxiety about whether to refer to the program by “he” or “she”. Commentators say they are tempted to use a gendered pronoun because the program’s behaviour reminds them so much of a person. The inevitable happened, of course: the choice to use “he” by at least one commentator was accompanied by a justification from sexist premises.

Our desire to use gendered pronouns will increase as we start to speak of AIs as if they have consciousness. We have an opportunity to do this right, and make a reasoned choice, rather than a thoughtless stumble. I therefore have a proposal, but it requires a change in English grammar. Our world may be turned upside down by AIs anyway, so why not change grammar while we are at it?

My proposal: That the gender of pronouns used for an AI is that of the person speaking about it. I would call AlphaGo “he”; my daughter would call AlphaGo “she”. Whatever is the third person pronoun that you prefer to be used for yourself (“he”, “she”, “they”, whatever), that is what you would use in speaking of the AI. A group of people speaking would give the AI a plural pronoun.

This solves the obvious problem of attributing a gender to the AI, but it also acknowledges that by being human, and a member of the world culture that created this AI, I am in some small way part of this AI.

That is my proposal. And while we’re at it, we can apply the same principle to God.

Thou shalt not use unbalanced parentheses for list item labels

Among the Commandments of my lab are:

Thou shalt not use unbalanced parentheses in numbered lists

I find lists like this an aesthetic and logical abomination: 1) first item, 2) second item, 3) third item.  The programmer in me cringes; the lover of symmetry cringes; and the lover of clarity cringes.  Why not include the left parenthesis, to treat the item labels as parenthetical remarks?  Why add a new meaning to the parenthesis, thus cluttering sentences with punctuational homonyms that need to be distinguished?  The aversion to this practice is shared through my family.

Now, on the website of the 2016 Evolution meetings is an example of why not to use these unbalanced parentheses:

Those interested in speaking in one of the Spotlight sessions below should submit an application to the organizer of that session (emails below in session descriptions). The application must include: 1) name, 2) institution, 3) rank (e.g., student, pdf, faculty), 4) names & institutions of co-authors, 5) a short abstract (max 300 words), and 6) which talk types you are willing to give (28, 13, or 5; the more flexible you are the more likely you are to be selected). Applications must be received by April 30, 2016. No financial support is provided to selected speakers. Additional details here.

Here, friendly paired parentheses framing side remarks, some containing numbers, are hobnobbing with their doppleganger asocial list delimiters. I find it difficult to scan.  This does not promote clarity.  The little lost parentheses are like fish bones getting stuck in my throat.

I need to go make some honeybush tea to calm down.


27000 images of jumping spiders

announcementImagesOver the years, I’ve tried to take photographs of living males and females of every species of jumping spider collected on most of our expeditions. This amounts to many hundreds of species, some of them poorly known or entirely undescribed. In total, I have about 27000 digital or digitized photographs of salticids, some going back to the 1970’s.

There has been no point in keeping them secret, but any system of organizing them for release was hampered by the lack of a good taxonomic organization of the family. Now that I have published a new salticid classification, that problem is solved. Thus, I am releasing most of my photos here: They are released under a Creative Commons license so that they may be re-used. My hope is that they will be useful to other arachnologists in their research, and that arachnological elves will incorporate them into Wikispecies and other places.

As noted, this collection includes many images of undescribed species. If you plan to describe some of these species, please contact me first, because I may already be preparing a species description. Consider this as a fair exchange, because if I am not describing the species, then you are welcome to use the photographs in your own publication describing the species.

A new phylogenetic classification of jumping spiders

announcementYesterday, my paper on the classification of jumping spiders was released online; the paper publication is 25 November. It’s good old-fashioned taxonomy/systematics, finished with hand-carved wood, and leather, and brass, but on the inside is phylogenetics, the union of my 44 years of looking at spider bodies and our 20 years of molecular phylogenetic work. It is the first new complete classification of the family published since 1903, and implicitly the first phylogenetic treatment of all 600-plus genera. The paper is available online here.

I expect that, among all the works in my career in empirical, theoretical and computational systematics, this paper will give me the most pride, and will best give the spiders the honour they deserve.

The new Andean jumping spider genus Urupuyu and a revised classification of the Amycoida


Urupuyu edwardsi, male. © 2015 W. Maddison, Creative Commons Attribution (CC-BY) 3.0 license

In the cloud forests of Ecuador we found, in 2004 and 2010, some small shiny black jumping spiders that we couldn’t recognize. Gustavo Ruiz and I have just published a paper describing them as the new amycoid genus Urupuyu, from the Quechua words for spider (uru) and cloud (puyu). This paper does a lot more than that, however. We take the opportunity to do a phylogenetic analysis of the whole clade of amycoids, and present, finally, a comprehensive phylogenetic classification of the group. I’m proud of what Gustavo and I accomplished with this paper, as it is a major step forward for salticids in the Americas.

Ruiz GRS, Maddison WP. 2015. The new Andean jumping spider genus Urupuyu and its placement within a revised classification of the Amycoida (Araneae: Salticidae). Zootaxa 4040: 251–279.

Hurricane Patricia descending on Chamela

Early in 2014 I spent several wonderful weeks at the Chamela field station of the Universidad Nacional Autónoma de México, along the coast of Jalisco just SE of Puerto Vallarta.  The habitat is a rich tropical deciduous forest, open and dry in some seasons, wet and green in others.  The field station is excellent, well run, with good infrastructure.  And now, the strongest hurricane ever recorded in the Western Hemisphere is projected to hit the coast of Jalisco directly at Chamela.  The good news is that the station is a kilometre or so away from the coast, and on a hill.  But I imagine that there will be considerable damage, at least to the forest.  I hope the people and their ongoing long-term studies will be OK.