A related, though much closer to home, experience I have had is an airplane flying through the field of view during a reasonably long exposure on a galaxy. The plane's transit left two long, continuous streaks through the image, one from each navigation light on the wing tips. The streaks had different brightness, likely due to the transmission of the filter I was using for red and green light. So in principle you could use knowledge of the filter transmission curve to work out which direction the airplane flew through the camera's field of view.
Well you don't need to do that, because the left wing has a red light, right has green. So as long as you're comfortable with the very reasonable assumption that the plane isn't upside down, you know which direction it was going.
Edit: sorry, I didn't mean to say that you can't do what you suggested, more that if you were solely interested in direction you wouldn't need to. It'd still be an interesting exercise!
> Well you don't need to do that, because the left wing has a red light, right has green. So as long as you're comfortable with the very reasonable assumption that the plane isn't upside down, you know which direction it was going.
The overwhelming majority of astronomical images (i.e., those not taken with a DSLR or other consumer-like camera) are monochrome images. So in any single exposure you don't retrieve a RGB color image. The color images you see from all professional and many amateur telescopes are reconstructed from multiple exposures through different filters (e.g., the one in the syfy.com article). So you would not see a red light and a green light – you just see two monochrome streaks with different brightness. Planes pass through a telescope field of view relatively quickly, so you most likely only see it in a single exposure.
The way that they determined the the object was moving, and the direction, by comparing consecutive images is exactly the way that scientists are currently looking for a predicted dwarf planet quite far out in the solar system.
If you were unanaware, most telescopes have data available online these days, provided their data is also public and not embargoed. There's REST API you basically do something like:
POST /tap/sync?query=select * from object a where ... HTTP/1.1
Yes, there's super-subset of SQL-92 called ADQL which is ran, but no inserts/deletes/updates are allowed. You can upload data to the services as XML files.
I didn't write the API, I just bitch about how it should be improved, but it really is a REST api.
You post to a query resources, which creates a query-resource object for either sync or async queries. If it's an async query, you get a query id back, which you can use to query the status of... your query. When it's done, you can get back the results.
I never said anything about JSON. REST (in HTTP's case) is about a) identifying resources (instead of, say, operations) by URIs and b) effecting state changes on those resources using HTTP verbs.
HTTP/1.1 GET /foo/deleteWidget.asp?id=1234
is not REST (but such things were and are still common)
HTTP/1.1 DELETE /foo/widgets/1234
is.
If a Web service where you can execute database queries by passing raw SQL as a GET parameter (I really hope the database has its permissions set properly, otherwise holy SQL injection, Batman!) is a REST API then every Web service that has ever existed is a REST API and the term is completely meaningless.
"every Web service that has ever existed is a REST API and the term is completely meaningless."
Yes, the term is completely meaningless. In the strictest sense that some people insist on there appears to be 0 "real" REST APIs in the world. In the loose sense, as you say, everything is a REST API. I've really never seen a definition of "REST API" that usefully splits the world of HTTP-based APIs into "REST" and "not REST" that anyone is willing to stick with; the term seems to beg for people to start a purity spiral that ends with whatever supposed REST API was being held up as an example of a REST API turning out to violate some tenant or other of "real REST".
I'm seeing a strange white dot float at the exact center of the browser on the page. At first, I thought the page was being cute, with an "astronomical artifact" of dust in the lens, but the article didn't mention it.
Just recently went to the Goldendale Observatory in southern Washington, and man was that fun. We got to look through their giant amateur telescope, as well as do some naked eye spotting.
Go on a night when there's little moon (quarter phase?) and no weather. It's a lot of fun, and the people working there are very charismatic.
Apart from "that, my friends, is science!", the follow-up he got is another nice example of https://www.xkcd.com/386/ — a.k.a. "Someone's wrong on the Internet"
I don't think the XKCD really applies here, at least in my interpretation. This was more like a mystery that everyone worked together to solve rather productively.
To me, the XKCD represents pointlessly arguing over random stuff while not making any productive points. Come to think of it, perhaps I'm falling foul of it again as I type...
Personally, I always had (and still have) a much more positive view of the XKCD. I believe it actually shows an interesting way how people try to make the world better, and share their knowledge, in face of noticed errors.
I believe this mechanism can be also used with premeditation, as an interesting way to encourage collaboration and solicit help. I seem to recall, that in early days of Wikipedia, it was explicitly suggested that editors should not fix small errors and typos in articles, so as to provoke newcomers to fix them, and this way "interactively" teach them how wiki works and provoke ("positively trick") them to start collaborating. Based on this, I'm sometimes pondering whether conscious erroneous assertions (e.g in a blog article) might possibly be a better way to solicit help on the Internet (e.g. via blog comments) than explicit questions. But I'm not quite sure, and also I can't shake the feeling that it'd be manipulative.
