![]() |
![]() |
![]() |
![]() |
![]() |
I have some things collecting in tabs, so here's a hodge-podge:
First, border-crossing. You've probably heard by now that border control in the US has gotten aggressive, including demanding passwords for encrypted devices and then taking them out of view for an extended time. You don't have to give up your password, but if you don't, they can confiscate your device for weeks or months for "review". The rights you have against unreasonable search and seizure in the US are not the same as those you have at the border. While they can't deny entry to US citizens, they can to others.
It's important to know what you don't know. David Director Friedman has an interesting idea about applying economics to teaching -- specifically, grading exams.
A lot of the Rands article The New Manager Death-Spiral sounds very familiar.
We all know one of the Internet rules: don't read the comments. The parts of the net I frequent tend to be better than, say, a random sample of YouTube, which is due to a mix of conscientious participants and comment moderation. A while back I came across a comment-moderation policy described as "Victorian Sufi Buddha Lite": they require a comment to be at least two of true, necessary, and kind.
I'm not sure that philosophy applies to windshield notes, but they sure are funny.
How do we know Humpty Dumpty is an egg? The rhyme doesn't say so. Huh, I never thought about that.
Speaking of things I hadn't thought about, have you ever noticed the similarities between fantasy-adventurer settings and westerns?
The rabbis are discussing things that are and are not automatically included in a sale. If a man sells a house the sale includes the door, a mortar fixed in the ground, and the casing of a handmill, but does not by default include the key, a movable mortar, or the sieve used with the mill. If he sells a courtyard the sale includes the houses, pits, ditches, and caves attached to it, but not movables. (R' Eliezer disagrees, saying he sells only the courtyard.) If he sells an olive press, the sale includes certain parts but not others. (My knowledge of olive presses is inadequate.) In all of these cases, however, if he says "I sell you this (whatever) and all its contents", the sale includes all those items that would otherwise have been excluded. (65a-b, 67a-b)
When I bought my first house I learned that if the contract doesn't specify certain things, they're not included -- even if they seem to "go" with the house, like window blinds. This seems to be like that. (Keys are automatically included in my experience, though.)
Originally posted at http://cellio.dreamwidth.org/2000918.html.The mishna turns to laws of privacy. If two houses share a courtyard, the owner of one may not place a door facing the other's door or a window facing the other's window, nor may he enlarge ones that are already present. On the side of the street, however, he may do so. Where do we learn the rule for privacy? The g'mara tells us: from the story of Bilaam, who lifted up his eyes and saw Israel dwelling according to their tribes. This indicates (the g'mara says) that he saw that their doors did not exactly face each other, and this caused him to say: worthy are these that the Divine presence should rest upon them! Why then is there an exception when the houses are across a street? Because the one can say to the other: you have to preserve your privacy from the eyes of those on the street anyway, so you might as well from me too. (60a)
I expected the "across the street" argument to be about distance. Here the argument is "I'm not making it any worse for you".
Originally posted at http://cellio.dreamwidth.org/2000439.html.
Earlier in this tractate the talmud discussed that it takes three years to take possession of someone else's property by occupying it. Now the mishna tells us: this rule of three years applies only to occupiers, but not to recipients of a gift or an inheritance (or those taking the property of a deceased proseylte, which is more complicated and I'm hand-waving it away today). The g'mara explains that the rule of three years applies when the acquirer has to make a plea -- if the original owner says "I did not sell it" and the occupier says "I bought it", that's an example of needing to make a plea and thus requiring three years. But a claim requires no plea in the case of a gift or an inheritance;, and all the occupier has to do to establish his claim is to start to build a fence or a door. (A modern equivalent might be changing the locks.) The g'mara then goes on to discuss how much he has to build in order for it to count. (52b)
Today's daf is 53.
Originally posted at http://cellio.dreamwidth.org/2017/03/16/bava-batra-52.html.Stopping GAN Violence: Generative Unadversarial Networks
Samuel Albanie, Sébastien Ehrhardt, João F. Henriques
While the costs of human violence have attracted a great deal of attention from the research community, the effects of the network-on-network (NoN) violence popularised by Generative Adversarial Networks have yet to be addressed. In this work, we quantify the financial, social, spiritual, cultural, grammatical and dermatological impact of this aggression and address the issue by proposing a more peaceful approach which we term Generative Unadversarial Networks (GUNs). Under this framework, we simultaneously train two models: a generator G that does its best to capture whichever data distribution it feels it can manage, and a motivator M that helps G to achieve its dream. Fighting is strictly verboten and both models evolve by learning to respect their differences. The framework is both theoretically and electrically grounded in game theory, and can be viewed as a winner-shares-all two-player game in which both players work as a team to achieve the best score. Experiments show that by working in harmony, the proposed model is able to claim both the moral and log-likelihood high ground. Our work builds on a rich history of carefully argued position-papers, published as anonymous YouTube comments, which prove that the optimal solution to NoN violence is more GUNs.
The talmud is (still) discussing adverse possession of property, chazakah. Last time we talked about real-estate; now we're talking about other property. We learn in the mishna that a craftsman (making repairs, for example) has no chazakah, and the g'mara discusses this at some length. Rabbah argues that this means only the case where the owner delivered the item to the craftsman in the presence of witnesses. If, instead, he delivered the item without witnesses, then the craftsman can say without fear of contradiction that he never had it or that somebody else gave this item to him. Or he might make the more probable claim that he (the craftsman) purchased the item, and in this case we believe him. According to Rabbah, determination of the owner's claim depends entirely on whether the owner has witnesses to the original transaction. Others, however, will argue that he has to see his item in the craftsman's hands; Rabbah would say that if he delivered it and the craftsman then hid or disposed of it, the owner has a claim. Others would say he does not have a claim if he hasn't actually seen it. (45a-b)
Today's daf is 46.
Originally posted at http://cellio.dreamwidth.org/2017/03/09/bava-batra-45.html.The talmud is discussing hazakah, adverse possession of property. The general principle is that if you occupy somebody else's property long enough and the owner doesn't protest, you can claim ownership. The g'mara records a dispute about whether the owner has to make his protest directly to the occupier (as opposed, say, to going to a court). It records one argument that says that taking possession of a fugitive's property does not confer hazakah, and this is because the fugitive can't very well return to make his protest. But another says you can possess a fugitive's property.
Another argument says that the reason we allow three years for the objection if the owner is overseas is that we need to allow time for him to first hear about it and then return; if he could make his protest from where he is, we wouldn't need to allow time for him to return. Raba says the law is that you do not need to protest in the presence of the occupier (that part about returning from overseas is just good advice), because "your friend has a friend and your friend's friend has a friend" -- in other words, we're relying on the rumor mill to get word to the occupier that the owner has protested.
And what is a valid protest? If he says "So-and-so is a robber", this is not sufficient. If he says "So-and-so is a robber who has seized my land wrongfully and I'm going to sue him tomorrow", that's a valid protest. He must make this protest in front of either two or three people (there's a dispute about that), and he must not bar them from spreading the word. (38b-39a)
Originally posted at http://cellio.dreamwidth.org/2017/03/02/bava-batra-39.html.It's the season of Purim Torah on Mi Yodeya. Here are some of my favorites:
From this year:
My answer to a question asking what Birkat Amazon is and what giving thanks after a meal has to do with amazon.com
What are we supposed to do about losing an hour of Purim this year? (I answered that one too)
Why do we shake four heretics on Sukkot? (on Sukkot we shake the "four species", different plants; the word for "species" resembles the word for heretics)
And some from past years:
And one I had a lot of fun answering, Why don't Jews accept Our Lord and Savior?
There are a lot more where those came from, and the season continues for about the next two weeks.
Originally posted at http://cellio.dreamwidth.org/2017/02/28/purim-torah.html.

The g'mara on today's daf discusses two cases of disputed ownership. In the first case, a man said to another: "what are you doing on my land?" The other said "I bought it from you; here's the deed of sale". The first said it's a forgery; the second then said to Rabbah "yes it's a forgery; I had a real deed and lost it". Rabbah rules in favor of the man occupying the land, saying why would he lie? He could have claimed the deed is genuine; since he instead told us this story about having had a deed, we accept his lesser claim. But R' Yosef objected, saying it's mere clay (and he's admitted it!). Rabbah wins this round.
In the second case, a man said to another: "pay me the hundred zuz you owe me; here's the bond". The second says "that's forged". The first again told Rabbah that yes it's a forgery but he had a real one before. Rabbah again says "why would he lie?". And R' Yosef again says it's mere clay. R' Yosef wins this round.
What's the difference between the two cases? R' Idi b. Amin says in the land we follow Rabbah because we say "let the land remain in its present ownership", and in the money we follow R' Yosef because we also say "let the money remain in its present ownership". It's not about land verses money; possession, according to this g'mara, determines the outcome absent proof. (32b)
Originally posted at http://cellio.dreamwidth.org/2017/02/23/bava-batra-32.html.Can you believe that I've been online since the ARPAnet and yet, in 2017, do not know the nuts and bolts of domain-name management? Perhaps you, dear reader, will point me toward the clues, and I promise not to be offended that you're quietly snickering there.
The recent LJ upheaval is far from the first signal that really, if you care about durable links, you need to own your own domain, but it's the one that's finally gotten through to me. Instead of relying on Livejournal or Dreamwidth or Medium or whomever else to provide a durable path to my stuff, I ought to control that, so if a service goes belly-up, the old, public URLs still work (with content migrated elsewhere).
What (I think) I would like (please tell me if this is flawed): some domain -- I'll use cellio.org as my example, though that one is taken so I'll need another -- where www.cellio.org points to my ISP-provided web content (which I can easily edit), blog.cellio.org points to my DW journal, medium.cellio.org points to my Medium page, and I can set up other redirects like that as needed. So I can't do anything about links that are already out there, but I can give out better URLs for future stuff (stop the bleeding, in other words). Bonus points if the durable URL stays in the URL bar instead of being rewritten (unlike pobox.com redirects), but that might be hard.
I do not want to run my own web server.
Now I already pay pobox.com for, among things, URL redirection, but it's to a single destination. And it's not at the domain level -- www.pobox.com/~cellio redirects to my ISP-provided web space. It'd be ok, though a little kludgy, if I could manufacture URLs like www.pobox.com/~cellio/blog that do what I described above, but unless there's something I can drop into my own web space, without requiring access to my ISP's web server, I don't think I can do that. Also, this leaves me dependent on pobox.com; owning my own domain sounds like a better idea. pobox.com has been solidly reliable for 20+ years, but what about the next 20?
I understand that I need to (a) buy a domain and (b) host it somewhere, and if I were running my own server then (b) would apparently be straightforward, but I don't know how to do that in this world of distributed stuff and redirects. Also, I'm not really clear on how to do (a) correctly (reliably, at reasonable cost, etc).
So, err, is this a reasonable thing to want to do and, if so, what should I do to make it happen?
Originally posted at http://cellio.dreamwidth.org/2017/02/19/domains.html.Me: What do we need, that is available from Amazon, that costs $4.02?
Dani: Is this what's known as a first-world problem?
Me: Not the most egregious I've seen, but yeah.
Free shipping: modifying buyer behavior since...whenever they started that. But it works because of their enormous catalogue; you can find something to fill out an order. (For personal orders this is pretty much never a problem, but I was buying house stuff and thus using shared money.)
(This post is a minimal test of Dreamwidth's Markdown support. Let's see what happens.)
Originally posted at http://cellio.dreamwidth.org/1997794.html.The g'mara then discusses R' Akiva's restriction. When he makes an exception for the west, is he saying that you can build closer there (the 50-cubit restriction applies only to the other directions), or is he saying you can't build there at all? The latter: R' Akiva says that the west is a "constant abode". Well, that just raises more questions -- a constant abode for what?
Shall we say it is the constant abode of winds? No, because there are four winds. No, it's the constant abode of the Shekhina (the divine presence), as it says in Nechemiah 9:6: "and the host of heaven worships You" -- meaning that the sun and the moon in the east bow down to the Shekhina in the west. R' Aha bar Yaakov objects, saying that this means when the sun and moon are setting, i.e. are in the west, so they bow to the Shekhina in the east. R' Oshaya objects to both of them, saying the Shekhina is everywhere, and R' Sheshet agrees with him. R' Abbahu then argues that the Shekhina is in the west because 'Uryah, a word meaning "evening", is equivalent to avir Yah, meaning "air of God". (25a)
Not addressed here: what about the next town to the east? Fine, I place a tannery to the west of my town to avoid putting it in my west, but haven't I then put it in theirs (if there's a town there)? And is it really unlimited distance, or is it more "out of sight (smell), out of mind"?
Also, the sun and moon might bow toward the west (according to some), but when we pray we humans bow toward Jerusalem because that is the "divine seat", so to speak. For me, that's eastward (and south some).
Originally posted at http://cellio.dreamwidth.org/1997529.html. comments there. Reply there (preferred) or here.
Originally posted at http://cellio.dreamwidth.org/1996707.html. comments there. Reply there (preferred) or here.
Originally posted at http://cellio.dreamwidth.org/1996223.html. comments there. Reply there (preferred) or here.
If you enjoy analyzing data about the Stack Exchange network, 2016 was your year. Community member Monica Cellio wrote a tutorial about our data explorer, which is maintained by another community member, Tim Stone. A resident data scientist, David Robinson, released StackLite, a lightweight version of community data. To see it in action, consider scripting language trends on Stack Overflow. In December, we connected our data to Google's BigQuery. People are already finding interesting results. Our data team has been posting analysis on the blog, if you crave more.(There are other links in that paragraph that I didn't recreate in this post. You'll need to go to their blog post to get those.)
The link in the quote is to the blog post where they announced the tutorial (in June):
Have you ever wanted to get a statistic about your favorite Stack Exchange site, but been baffled by exactly how to do that? The Stack Exchange Data Explorer (SEDE) may be just what you're looking for. SEDE was created to make it easy to query and browse the public data that we release periodically for all Stack Exchange sites, but a lack of familiarity with SQL has been a barrier to many of you who would otherwise benefit from it. Now, thanks to friend of the company and moderator extraordinare Monica Cellio, you have a tutorial to guide you in using it!Originally posted at http://cellio.dreamwidth.org/1995955.html.[...] But even though SEDE is nicer to work with than a raw data dump, it can still be pretty intimidating to new users, especially those who aren't trained database engineers. Up until now, the Data Explorer's own help docs have been a little thin, and mostly covered very specialized, advanced features. We've wanted to expand the guidance there for a while, but never quite got around to it. Then Monica rewarded our procrastination by helpfully volunteering to take on the writing.
Bava Batra begins by talking about fences. If joint owners of a courtyard (that is, the owners of the houses bordering it) agree to build a wall, they should make it down the middle and each should contribute half of the building materials. If in the future the wall falls, it is assumed that the place it occupied and the stones/bricks belong to both equally. In a place where fences are not customary, like in a cornfield, one who wants to build a fence cannot compel the other to share in the effort. If they don't agree, the one who wants the fence must build it set back from the property line, making a facing on the other side. If the wall later falls, we assume that it will all still be on his property and it will be clear that the land and the materials belong to him. (2a)
Today's daf is 4, and it's in the middle of the g'mara discussing this mishna.
Originally posted at http://cellio.dreamwidth.org/1995750.html. comments there. Reply there (preferred) or here.
Now on today's daf we learn: a man may not take a pledge from a widow at all, whether she is rich or poor. That is the mishna, but R' Shimon says in the g'mara that he can take a pledge from a rich widow but not a poor one. This is because you must return the poor person's pledge each night (for a garment). That's ok in the case of a male debtor, but if a man shows up at a widow's house every night to return her pledge, that looks bad in front of the neighbors, who won't know it's to return a pledge and just see a man visiting every night. Therefore we just don't allow taking the pledge in the first place for a poor widow, but a rich widow wouldn't need to have the pledge returned each night so that's ok. (113a, 115a)
The g'mara here does not consider changing circumstances. I speculate that R' Shimon would require the man to return the pledge (and then not retake it) if the rich widow had a change of circumstances.
Originally posted at http://cellio.dreamwidth.org/1995377.html. comments there. Reply there (preferred) or here.