Every once in a while, people tend to doubt how devious journalists can be, and make fatal mistakes like underestimating them. Today’s story is about plagiarism, devious journalists, and what the consequences of actions that people are sure they can get away with are. It may be a bit long, but trust us, the punchline is worth it.
Our tale started out with the good but devious folk at SemiAccurate getting sick and tired of being ripped off every time we get a good story. Some say that is part of the game, but it is not, it is fraud. You need to credit your sources, where applicable, or you can find yourself in deep deep trouble.
Some publications are ignorant about what plagiarism is, others are not. More of them claim ignorance, or have some very tenuous reasoning when called on their misdeeds. Some simply deny it, others make it a business practice, and most just get away with it. Or think they did, but more on that in a bit.
All it takes is a quick read of plagiarism.org specifically this page, to see how wrong they are. In the academic world, plagiarism will get you thrown out of an institution in short order. In the corporate world, if the company is worth a damn, it will get you fired, or at least severely reprimanded. If the company isn’t worth a damn, it is sadly business as usual.
Back to tech and journalism where plagiarism is rampant. When a story comes out that looks eerily like another, there are three possibilities. The most common is that one is the work of another, and is properly credited, it usually just takes a link. On the other extreme, you can have two stories that are indeed entirely coincidental, but this is a vanishingly rare circumstance. The best way to differentiate between coincidence and plagiarism is how different the stories are, especially in the details.
Unfortunately, the rest are usually plagiarism, and worse yet, most people tend to think they got away with something if they are not called on it repeatedly and publicly. This is the subject of our story, and getting back to the devious part, what we at SemiAccurate are doing about it. No, not just calling people on things, that is the easy part, but making sure that we can identify the practice unequivocally.
After a little brainstorming session on the topic, the goodish people at SemiAccurate came up with a cunning plan. We thought of ideas like making up fake front page stories, but that has unintended consequences, both to us, our readers, and the poor PR folk who have to answer questions about things that aren’t real. And it just lacks class. That plan was shot down. For now.
In the end, two ideas emerged to identify the practice, the first is to change a small and insignificant number or two in a large chart. Good examples are to change a single CPU price from $226 to $227, or to subtract a day from a release range. Going a bit further, you can add details that don’t matter to anything, but sound really cool, like, “The move to a 7% niobium alloy upped the efficiency of the new Ivy Bridge heat spreader design, leading to a 4 degree cooler CPU”. The other was occasional name dropping.
If you are not simply ripping off the info, and are doing the work that you claim to be doing, or have the sources you claim to have, then you will either know the plants are complete bull, or have your own without the same ‘mistakes’. This is what most writers claim to do, but vanishingly few actually do. This is true even when it is just looking up a manufacturers web page or firing off an email asking for a spec list.
If you cite your sources, you have a get out of jail free card too. “We at site XYZ said ABC yesterday, but our source, site DEF contained an error.” Simple, and no harm there, it all goes back to the one who make the error. If you open the New York Times, you will find updates and corrections like this on a regular basis, it is the right way to do things.
If you do things the wrong way, just stealing info, you have to then defend yourself in two ways. First, you have to say why you got it wrong, and that is annoying enough. More problematic is telling people how you got the same obscure things wrong as someone else. It is one thing to have to journalists both told an incorrect clock speed from the same source, but it is entirely different when two sites have the same price list with the same two chips each wrong by $1.
Trust us, it gets ugly quick and is quite impossible to explain away, especially when you are wrong. Some sites try and spin, but always end up looking like idiots. Others try and destroy the evidence, usually in a clumsy panic, and that ends up going from idiocy to hilarity. Why? Because anyone who is going to call a site out will have screenshots and PDFs saved long before the plagiarist has any clue they are busted.
If you recall, we mentioned being devious earlier. That, coupled with a job that requires documentation and research skills, makes for a deadly combo. When someone does something really blatant, and it is flat out undeniable, that is where the fun begins. That is also, somewhat un-coincidently, the point of this story.
Last week, another plagiarism flap came up, and one of the last times we weighed in on it was an open letter to the folk at Kitguru. It was prompted by four things, starting with their quite solid story on the topic, they were one of the few who had the stones to weigh in on it. Kudos to them for that part. The open format was dictated by the site not having any author or editor contact info available, so it ended up as a story instead of an email. Then there was the simple fact that, given their decent analysis of the then current situation, the Kitguru take on the questions posed in the letter were interesting and relevant. Lastly, and most importantly, was the simple fact that they were dirty as hell.
Yes, the whole open letter was an intro to this piece. To date, we have had no response from Kitguru personnel on the letter, so we will do this without their input. Should they choose to respond, in writing, we would be more than happy to post their comments, but as of now, there aren’t any.
What are we talking about? On April 8, 2011, Kitguru’s author Jules, likely the same Jules who wrote the plagiarism story, wrote one titled, “Intel to launch Pelican Lake shrink on 7nm in 2016“. Nice juicy scoop there, 5+ year out code names are really hard to come by, especially those with architectural details. Stories like this are few and far between, trust us, we know, we break a lot of them. On top of that, there are a lot of juicy details in this story like, “Technologically, Pelican Lake is streets ahead of Sandy Bridge, with levels of parallelism that present designs can only dream of.” Wow, cool.
Luckily, Jules is not the only one to have this information, Faith, another Kitguru person, says in the comments, “but this is the info we have from Intel.”, and, “Our source was very specific about the 7nm update, but there does seem to be a ‘hole’ around 10/11nm”. Just in case anyone doubts the independence of Kitguru’s sources, Faith responds to the open letter in their forums, “There are many people in this industry who have been around a while and various groups do talk, but this one did come up randomly.” They sound quite sure of their deep throats at Intel, but do dodge the question posed quite artfully.
One thing that they never do is credit or source SemiAccurate’s Copper who first mentioned Pelican Lake in a forum post in December 2010. That is why we asked the questions in the open letter to Kitguru. Is it coincidence? Did they not know about what Copper wrote? How did they manage to get such juicy details about the chip? No matter what the answers are, there is one problem that they can’t explain away, the fact that we made the name up.
Yes, there is no Pelican Lake. Actually, there is a Pelican Lake, several in fact, but none at Intel. The one we were thinking about when the forum post was made was this one, it is a beautiful area in northern Minnesota where Copper went fishing as a child. We thought it would fit in well with the Intel naming schemes, and were laughing pretty hard when we wrote the post in question.
You may notice that the Kitguru story has a ‘gap’ after Haswell/Broadwell, just like the story linked by Copper. The intonation in the forum post is that the successor is Pelican Lake, but it isn’t, it is Sky Lake, then Skymont, then at the time, nothing. We knew these names at the time, and finally wrote them months later. The Pelican Lake thing was purposely written up to intone things but not state them. To quote the venerable Admiral Ackbar…. And Kitguru walked right in to it. Blatantly. Badly. Stupidly. Hilariously. For us anyway, not so much for Kitguru.
What more is there to say? Well, plagiarism is a bad thing, don’t do it. Cite your sources, they are a get out of jail free card as well as being the right thing to do. We weren’t joking when we said that, it would have saved Kitguru a lot of embarrassment, we really look forward to the explanation here. They have good sources, do have good material, and usually cite correctly unlike many ‘reputable’ sites. All it takes is one ‘whoopsie’, to destroy a reputation. See above for more details.
The moral of the story? Plagiarism is wrong, period. Plagiarism is a bad idea, especially when you are doing it to devious people that are 17 steps ahead of you in the game being played. Stealing from people who set mines for a living is a bad idea too. We have a lot of mines placed, and there is a long list of web sites that are somewhere between the click and the bang right now. Cite your sources kiddies, it may prevent you from being “Kitguru’d”.S|A