Tuesday, October 17, 2017

me too?

It's almost certainly idiotic of me to chime in at all, and my normal policy is to avoid gender issues at all costs, but I'm apparently feeling stupid today... 

What do people mean when they say "me too"? 

Yes, I've been catcalled and I've been creepily followed. I've been aggressively hit on to the extent I was concerned for my safety on many occasions, most recently last month. I've had to physically resist men who didn't take no for an answer. I have only compassion for those who have experienced worse.

On the other hand, every male who has ever been in a position of authority over me has been very overtly very concerned with not doing or saying anything that might possibly be interpreted as harassment*. When I quit the one real job I ever had I had at least four meetings with different people terrified that I would accuse someone of harassment (while nothing could be further from the truth). In my current job I have senior male colleagues who actively look out for situations in which women are inadvertently being taken advantage of. And people say economics is particularly bad for women! The only time I've regretted being a woman in economics is when I have to be on committees because it's required that there be at least one woman on every committee...

Maybe I'm exceptionally lucky, but I believe I'm in the first generation of women for whom this is generally true, and I believe that's amazing progress. Assholes will always be assholes, but they're a small and inevitable minority.

(Being an extreme introvert, now that's something I wish I could change...)

~~~

* I particularly like to tell the story of how uncomfortable my mentor looked when telling me, at age 16, about the source extraction software called "sextractor".

Wednesday, October 11, 2017

Mendeley+Elsevier

In case anyone else is googling around about the same problem, or just wants to hear a funny story about the joys of cloud* services...

Part 1: Elsevier has a centralized editorial website for all their journals that you can log into as either an author, a reviewer, or an editor. This is a recent agglomeration of many individual websites for individual journals, so you can link those individual journals to your central account as you encounter them. This works ok, but then Chrome remembers my old passwords for the specific journals, and I paranoidly question whether I've set up a different password for reviewer login versus author login, so the end result is I've reset my password with them many times. I always reset it to the same password though, so the rate has been falling steadily.

Recently though, I haven't added any new journals to my account, and I've still had to change the password several times in the last two weeks. I used the saved Chrome version, verified in Chrome's password manager that it is in fact the password I remember setting, reset the password to exactly the same thing, and then it works fine. I feel like I'm going crazy, but hey, never underestimate the ability of the cloud to screw up basic things.

Part 2: For organizing papers, I use Mendeley desktop. Mendeley is buggy as a bog, which for a long time I was forgiving of because it was a brand new and free service, but now it's been 10 years and I pay them to keep my papers synced across my work and home computers, so my patience is waning. But at least I've got my workflow set up, I know the bug workarounds I need, and it runs in linux well-ish, so so far the switching costs have kept me there (but if anyone has tested alternatives please let me know...) Anyway, many of the bugs are related to network access and syncing, and in the process of unplugging-and-replugging as many things as I can think of in an attempt to fix the situation, I end up resetting my password a lot. Always to the same thing, but still.

Recently though, and in an unfortunate coincidence with updating mendeley versions on my two computers so that I thought the upgrade was related to the problem, I've been forced to reset my password nearly everytime I open it on my desktop or try to log in online. Again, I check that the saved password is what I remember, reset it to the same one, and then it works.

Resolution: I couldn't find anything helpful on google and since I am a paying Mendeley customer I was optimistic they would actually respond, so I wrote off the Elsevier problem as a lost cause and emailed Mendeley support. And amazingly enough within 24 hours I had a useful response (addressed to "Dear Joel", but oh well, details.....) Apparently when the login page on the Mendeley website says "Mendeley now supports signing in with your Elsevier credentials." what it really means is "You MUST now sign in with your Elsevier credentials, and when you change your password here it will also change it with Elsevier". There is of course absolutely no message of the sort in the desktop client at all, which is what I use for everything except resetting the password. So I've apparently been resetting both passwords simultaneously, alternating between the two.

I don't remember actively linking my Mendeley and Elsevier accounts, but I'll acknowledge the possibility that I've just forgotten or didn't realize that that's what I was doing. If so, I apologize for complaining about this particular issue and instead ask you to redirect my complaints towards the fact that Mendeley crashes on me several times a day and has to be restarted every time I ask it to sync something and gives me meaningless ungoogleable connection errors randomly when I open it and won't let me just skip the damn login process and see my local database while it ponders whether to connect this time around. And as for Elsevier, well we all know what else we can complain about with them...

~~~~

* The cloud is one of things about which I drift pretty close to the fearmongers' warnings about impending dystopia. I'm not sure who came up with the idea, after seeing how Web2.0 royally fubared the web, to put everything on it. So use syncthing instead of dropbox, back up your gmail (gmvault works well), and use distributed/opensource/offline/encrypted/whateverelse systems when possible.**

**And linux and emacs just because it will make your life so much better. And road bikes.

Tuesday, October 10, 2017

Behavioral Economics get the Nobel

Many congratulations to Thaler!

This is a prize recognizing behavioral economics as a field, and there are of course many other deserving pioneers (a couple of whom have already won prizes for slightly different things). No matter whose name is singled out, I'm happy the field as a whole is being recognized.

I won't add more commentary now; the blogosphere has it thoroughly covered. Maybe another time I'll write about why I simultaneously love and hate nudges. (Hint: It's for approximately the opposite reason George Loewenstein does.)