5 ms·
Google (and by extension Jeff Dean) have made their greatest contribution in that they successfully popularized niche computer science with decent implementatio
by jandrewrogers 13y ago
Google (and by extension Jeff Dean) have made their greatest contribution in that they successfully popularized niche computer science with decent implementations.
I think a point that is often lost is that Google invents very little computer science; they take obscure computer science that in most cases has already been implemented somewhere and do a very good, high visibility implementation of that computer science. They may apply it to a somewhat different problem domain but the solution already existed for the taking. Core high-scale computer science domains like HPC traditionally don't publish shiny geek-friendly know-how but there are very deep existing pools of expertise out there.
Google has excelled at branding concepts like MapReduce, Spanner, etc that already existed in real systems for many years. Google has made people aware of these technologies but it is a stretch to say they "invented" them in any material sense beyond publicizing their own implementations. I think they receive a little too much credit in many cases for invention when their primary contribution has been popularization.
- karterk 13y agoI think they receive a little too much credit in many cases for invention when their primary contribution has been popularization. That's being a little pedantic. Applying X for Y is definitely being inventive. That's how largely every domain works. We stand on the shoulders of giants etc.
- jandrewrogers 13y agoThe weakness of this contribution is often cited as the reason many types of software patents should be invalid. Putting a shiny wrapper on an old idea in an obvious but slightly different context is something upon which many software hackers would normally cast a jaundiced eye. My point was largely that e.g. MapReduce had literally been used in the HPC world for quite some time in countless variations. Something similar occurred with Spanner where it was a minor variant implementation on a concept that existed in many systems decades old. I worked on systems that used both of these "inventions" in the 1990s; no one thought it was novel then either. When I think of "invention" in computer science, I assume some hard problem in computer science was solved. That is not the case for most of what Google is credited with in the realm of computer science. The absence of meaningful novelty in Google's systems strikes me as a good basis for being skeptical of claims of invention. Relative obscurity of computer science does not confer invention claims on the first to popularize it. People frequently attribute more computer science invention to me than is warranted (I do a lot of high-end algorithm work). I always try to make it clear that much of what they see as "invention" is their first exposure to obscure computer science that I happen to be familiar with and know how to exploit. Much of what is attributed to Google is similar in character; people conflate the source of their introduction with the inventor of the concept.
- chubot 13y agoCan you give a citation for the MapReduce and Spanner predecessors? I've seen similar papers that predate MapReduce (I'm curious if you will find the ones I have). But I've also noticed that most people who think MapReduce is old hat have never used it and/or misunderstand what it actually is. And if it was already so well known, there wouldn't be hundreds of CS researchers continuing to publish both theoretical and systems of papers about it. The MapReduce paper has 7992 citations so far, and the number continues to increase every year! http://scholar.google.com/citations?view_op=view_citation&hl=en&user=NMS69lQAAAAJ&citation_for_view=NMS69lQAAAAJ:BqipwSGYUEgC http://scholar.google.com/citations?view_op=view_citation&hl... What paper should they be citing instead? It's hard to believe that 7995 sets of authors would ignore prior research, especially since some of those authors are a decade or two older than the MapReduce authors.
- jandrewrogers 13y agoAs a general point, a lot of computer science is poorly documented or difficult to find in the literature but well-known among the practitioners that use it. A great example is the recent flurry of activity (including patent activity) around "hyperdimensional hashing" and similar concepts. The irony is that the concepts are so old that some of the original patents have expired but the work is sufficiently forgotten that academics are attempting to re-patent those ideas. I am familiar with it because those algorithms are de rigueur for some types of massively parallel systems (I've been designing parallel systems for ages). No one publishes papers on it though because it is not new to those people. A large part of it is that much of the original literature is lost in paper form and cannot be found electronically. I've been burned by this personally, having done work that I later discovered had been previously done in the 1980s by a respected academic but which is not cited in any modern literature; from the standpoint of modern literature, the work is completely forgotten. But it had been done and rare people exist that are familiar with it. So more specifically (I got back from vacation a couple hours ago, forgive the terseness): Spanner is an old, old idea in literature and a number of bespoke systems were built on the concept. Ancient transaction processing literature mentions it, I've worked with a few crusty mainframe environments that used variants of it. It was considered a stupid idea for decades because the hardware required for a reasonable implementation was outrageously expensive for most applications. Atomic clocks and satellite-delivered time codes don't cost what they used to. I used to have a transaction processing book from the 1980s that described a couple specific implementations that were quite close to Spanner (written by a couple guys with Indian surnames, wish I remember the book). It had been written off so long ago for inexpensive commodity systems that apparently it took Google to rediscover it but it never fell out of use. Similarly, I think a lot of people in the HPC world were surprised to discover that MapReduce was patentable. Not every problem in HPC looks like computational fluid dynamics (CFD), which is often the stereotype. If you are actually an academic then you should know better than anyone that the number of cites is misleading. The MapReduce model (though not under that name) was a standard design pattern for some non-CFD problems in HPC in every respect going back to at least the 1990s. The specific novelty claimed by MapReduce has never been clear to me. A lot of the (terrible) compilers for obscure HPC platforms back in the day transformed the code into a canonical MapReduce model. Maybe I am missing what makes Google's version unique?
- aaronbrethorst 13y agoThis is a wonderfully-executed middlebrow dismissal. I hope pg adds it to his detection corpus (https://news.ycombinator.com/item?id=4714217 https://news.ycombinator.com/item?id=4714217)
- Wilya 13y agoYeah. Let's teach the algorithm to hellban people who disagree that Google is awesome.
- aaronbrethorst 13y agoI think you misunderstand what I'm saying. I have lots of issues with Google, but dismissing what they've done because it is insufficiently new is just silly.
- chubot 13y agoThat's exactly what I was thinking, but I couldn't find the link. Too bad it's still the top comment.
- jandrewrogers 13y agoThe commentary was terse (hiking in the mountains, limited IP access) but accurate. I have designed interesting algorithms and software on large-scale database and HPC systems for much of my adult life. As a consequence, I am much more familiar with these domains than most casual programmers or academics in these fields. The important point was that ideas like MapReduce and Spanner are literally rehashes of very old ideas that never fell out of use in some domains. But they are so old that no one that uses those ideas would dream of publishing a paper on it because it has been in continuous use for ages. In most cases, a deep enough dive in the dead tree literature will often turn up examples of these in real implementations. I learned most of it from graybeards working on cool stuff when I was young, not from the literature. I am not exactly young but both MapReduce and Spanner predate my time in computer science. BigTable was actually a novel composition of existing but obscure bits of computer science so I give Google full credit for that. Google has shown an exceptional ability to exploit ignored or forgotten computer science. However, to the extent they exploit it there is often little invention beyond those that they borrow the ideas from. In that respect, I think the credit is somewhat misplaced with respect to invention.
- gaius 13y agoExactly. They're like Madonna, or more recently Lady Gaga, they create nothing, but they take themes from the underground and bring them to a mainstream audience (and in doing so become far richer than any of the creators).
- jamesaguilar 13y agoThis is one of those Poe's law moments for me. I can't tell if you're doing a parody or just believe a crazy thing.
- AYBABTME 13y agoThe work Jeff Dean seems to be doing is _excellent software engineering_. 'Software Engineering' here I would define as 'the application of computer science'. Which is not a far fetched definition, although many dismiss it when the topic is raised. He does that very well, applying computer science in an engineering perspective and creating very good/performant/useful products. There's a lot of credit in doing that well.
- calibraxis 13y agoWhat's odd to me is the focus on specific individuals. For one thing, why not Dean/Ghemawat facts, at least? I always hear they participate on an equal basis; or does Dean carry Ghemawat like Sherlock carries Watson? Furthermore, when I actually read things about such universally-held geniuses, the hype often contrasts sharply with the moderate, logical introspections of the geniuses themselves. (Not always, like I vaguely recall of Nietzsche; but most often in my personal experience.) Quite often, the hype even patronizingly dismisses the person's own claims as mere humility. (As with Hume's coverage of Newton in his history of England.) And it often beats up on people who quite reasonably had issues with the person (as I hear with Socrates' spouse Xanthippe). When one hears breathless accounts of someone's superpowers, a logical first reaction is, "Cool, how can I attain that?" I wish it were otherwise, but the hype is always misleading for this end. As far as I can tell, it's storytelling. The hype fits in a certain dramatic constraint, like with profit-seeking movies which naturally can't portray too pedestrian a reality. In Dean's case, one of my suspicions is that US "nerd" audiences are more able to vicariously fantasize themselves as a "Jeff Dean" than a "Sanjay Ghemawat". Which, if true, would say something about the lack of imagination and homogeneity of such fantasizers. (And maybe these people serve as mainstream characters everyone's supposed to be in awe of, but it just wouldn't fit within current society if everyone could follow their interests. Who'll serve you at the store and mop office floors?) Anyway, sorry for the long random wondering about silly topics; clearly I'm procrastinating.
- nostrademons 13y agoLargely because it started as an April Fools joke (by Kenton Varda of Protobuf fame, IIRC, who had been working on protobufs under the mentoring of Dean & Ghemawat at the time), and the joke would fall flat if it were Dean/Ghemawat facts. It just doesn't have the same ring to it. Then it was leaked to the press, and the press wants heroes, because that's who ordinary readers can identify with. There are plenty of publications (usually in scholar journals) where Dean & Ghemawat get equal billing. They aren't Wired, and they don't usually get to the top of Hacker News (or actually, sometimes they do).