6 ms·
Interview with Hillary Mason, Chief Scientist of Bit.ly
- agotterer 16y agoHillary was also the key note speaker at pycon this year. Her talk and slides - http://www.hilarymason.com/blog/conference-pycon-2011-keynote/ http://www.hilarymason.com/blog/conference-pycon-2011-keynot...
- crasshopper 16y agoShe also made a twitter bot of the Berkeley Calculator: @bc_l. Hott.
- uyhtghj 16y agoChief scientist? It's a site with a lookup table and a 301 redirect In that case since I'm resposnible for locking the door at night here then I'm officially "superhero ninja defender of intergalctic freedom"
- aditya 16y agoNegative much? Bitly has massive amounts of data on what people are clicking on, and there's some serious data crunching to be done to tune things like content recommendation, etc. Last month, Bitly saw 7 billion clicks on its links and currently supports 8500 white label partners.
- uyhtghj 16y agoSize doesn't always mean clever. I just plugged an HDMI cable into my TV all by myself, it shifts more than 7Billion bytes/SEC to the screen.
- mquander 16y agoThat's right, and I suppose you don't give a lick of credit to the Chief Artist inside your television sifting through that data and drawing the picture on your screen, do you.
- wulczer 16y agoJust think about the availability constraints such service needs to comply with. I can't remember a bit.ly outage, and an outage would make half the Internet unhappy. Same for spam filtering, automatic reporting using the "+" links etc. At some scale level keeping up any service becomes a nontrivial task.
- gaius 16y agoSo don't use it, we already have DNS. If anything "url shorteners" are a step backwards, at least www.whatever.com is human-readable. I simply don't believe that we have enough computing power to add another layer to visiting a website but not enough storage for everyones URLs. It's just a stupid fad. Here's the new one: http://arseh.at/ http://arseh.at/ (real URL)
- jbooth 16y agoYeah, but it's more than sending a few 301s or plugging a cable, a better analogy would be that if you wrote a sophisticated algorithm to mine patterns out of those bytes to the tune of 7 billion whatevers per whatever and produce interesting results, then you could call yourself a digital signal processing scientist. They're not just doing redirects, they're doing a lot of data science with exploring the statistics of what gets shared, clicked, etc.
- revorad 16y agoI can't believe this troll is getting upvoted. Pure vitriol from a freshly minted account. Use the flag button, not upvote.
- uyhtghj 16y agoI'm an astrophysicist at a world famous institution - hence the dummy account. I just feel strongly about a corporate CTO deciding to upgrade himself by calling himself "chief scientist" - mind you it's not as bad as how my colleagues in the School of Engineering feel when the man who connects your new washing machine calls himself an "engineer".
- 18pfsmt 16y agoI would assume this person comes from academia with a title like that, so I checked out her about page because of your comment. I found that she is a CS prof on sabbatical.
- prunedevir 16y agoI found the about page a bit strange. Why doesn't she list where she was a professor? (Googling turns up Johnson and Wales University.) Saying that her "background" is in machine learning makes it kinda seem like that's not her particular academic field; no problem with that -- I don't think you need a PhD in machine learning in order to do good machine learning or data science work -- but it just sounds like something's missing. Googling a bit more turns up some research on developing a "virtual Morocco"/virtual classrooms for Second Life, so I think her academic field was about digitizing education?
- Nate75Sanders 16y agoI probably went down the same time-wasting search hole you did. I did so because every time I've ever skimmed her blog, I've never seen anything interesting on it. Then I watched the PyCon talk linked above and it's full of "cute" things with no substance. Skimming through some other talk gave me the same feeling. So I started digging. In Fall 2008, she was part of Path101 (http://www.slideshare.net/path101/path-101-opportunity http://www.slideshare.net/path101/path-101-opportunity -- slide 3) and listed as having a PhD in Computer Science. But like prunedevir, I had to dig relatively hard to find what university she taught at. There's no CV on her blog. There are claims that she's on sabbatical, so this is REALLY strange. Availability of her information is _nothing_ like someone who is a researcher in machine learning with a PhD. So I dug more. As prunedevir points out, the university is Johnson and Wales. It's a technical college. And she was an _instructor_ there (google for the JWU catalog 05-06) for a few semesters before they finally upped her to assistant professor (JWU catalog 08-09, I think). And in the catalog, she's listed as having an ScM, NOT a PhD -- which makes a whole hell of a lot more sense. I found a few slide presentations that she put together (sorry, it's a few hours later and I can't re-dig up all those links to post here) and they were all about PHP, CSS, and other web development stuff, as well as some stuff about the use of technology in education. I think most of this was from 2006-2008. You'd think the _chief scientist_ of bit.ly doing machine learning work would have been doing something else during this time -- not looking at run-of-the-mill web dev. Her previous website (3greeneggs), archived by the Wayback Machine (http://waybackmachine.org/*/http://3greeneggs.com http://waybackmachine.org/*/http://3greeneggs.com) shows more stuff consistent with web dev, etc, and awfully soft for a chief scientist working on machine learning. Furthermore, here's a paged removed from her blog that Google Cache allows us to see: http://webcache.googleusercontent.com/search?q=cache:hfTTGEbIhQ0J:www.hilarymason.com/blog/%3Fs%3Dnmc+hilary+mason+%2Bnmc&cd=4&hl=en&ct=clnk&gl=us&source=www.google.com http://webcache.googleusercontent.com/search?q=cache:hfTTGEb... I'd probably delete it, too, if I were pulling the same act that she is. Also, side note: in the PyCon talk she mentions that she looked up some old manuscript on statistics and it was by "A. D. E. Moyver" or something close to that. My jaw dropped. Surely she knew what she was talking about, but that name sounds suspiciously like a horrible bastardization of Abraham de Moivre -- and it was. He was exactly the guy she was talking about. I fully realize this post comes off as hugely negative, but I hate seeing people get ahead because of repeated self-promotion and gaming the system. Also, the Path101 slides could have been juiced up by someone other than her, OR the JWU catalog may be wrong and she really does have a PhD, but I way doubt the latter, and there's almost no doubt, for the former, that she at least would have seen the slide deck before it went out. I wasn't going to post anything, but this whole thing left a really bad taste in my mouth and the truth tastes a lot better.
- Apocryphon 16y agoDisappointed they didn't mention the country code of their domain...
- kno 16y agoShuuuut Please dont let Gaddafi hear that, he may retaliate :)
- gthank 16y agoGo read the relevant Quora thread http://www.quora.com/What-will-happen-to-http-bit-ly-links-if-Gaddafi-shuts-down-the-Internet-in-Libya-due-to-protests http://www.quora.com/What-will-happen-to-http-bit-ly-links-i...
- MediaBehavior 16y agoSeems a shame that wikibon.org interview couldn't spell her name correctly. See: http://www.hilarymason.com http://www.hilarymason.com