The following is an open letter to Kitguru’s editors, whoever they may be, in relation to their views on plagiarism. Since the site, specifically writer Jules, decided to weigh in on the topic without a method to contact him/her/it/they, this goes out in the open from the start. Hopefully, some readers of both sites may find bits educational.
First of all, well done on both the exclusives, and on correctly and linking stories. I do agree with what appears to be the Kitguru site policy, as stated by Faith in comments #14 and #17 in this story. Part of these comments are, “He had stuff we did not have, so it moved away from pure research into quoting the work of others – so we flagged it and linked them – which is our way.”, and, “We feel it’s about the fundamental ‘treat others as you want to be treated’, so we link wherever possible”.
I am not 100% sure that Faith is a part of Kitguru, I can’t find an exact list, but it sure reads that way. If he/she/it/they are not a part of the site, feel free to ignore this letter. Not trying to change the subject, we will move back to the Kitguru policy of linking. The problem in question is that if the above is indeed the way things are run, why wasn’t SemiAccurate’s own Copper, forum mod and kitchen sink cleaner, credited in this story?
The reason I say that is because as far as I can tell, Copper was by far the first to utter the name Pelican Lake in reference to anything other than a few fishing holes in the rural US, here. True, it wasn’t in a front page article, nor were there any details, but it is still published with a name on it, and as far as I can tell, there aren’t any other sources.
So, that brings up the question of Kitguru’s moles. Do you really have moles with access to 5-year roadmaps at Intel? Ones that know the deep architectural secrets and plans that far out, or did you get them from us and forget to credit? If you did get it independently, how does Kitguru, editorially speaking, reconcile that information? One person breaking the news is fine, that much is quite easy to defend with a ‘secret source’, no one else has it, so there is no need to defend anything. You can credit or not, that is a different topic.
If some one else has it, and then a person tells you about it later, you are faced with three choices. First, you can out the source by name and credit them. Second, you can credit the first person to break the story even if you got it independently, it avoids nasty questions later. Third, there is the way that Kitguru appears to have done, just claim it was independently sourced, and no citation needed, and no sources burned. Fair enough, and finding the original source can be time consuming and problematic. Some purposely avoid it entirely.
As Faith again says in the bottom of the Pelican Lake story, “Our source was very specific about the 7nm update, but there does seem to be a ‘hole’ around 10/11nm”, and those would be Sky Lake and Skymont. Did both Jules and Copper both come up with Pelican Lake independently? How do you prove it if they did? One was clearly first, the other a bit more complete. Is a citation needed? Faith says in the same comment, “Not sure if this will format correctly, but this is the info we have from Intel.”, but what do you do when it goes to the next level like some other recent dust-ups?
With ‘secret sources’, you have to be unquestionably first, or face potentially annoying questions. You raise some of this in your article about BAPCO and related debates, but how does Kitguru deal with this? How do you deal with people who are, and I am not suggesting this is the case here, caught with their literary pants down? I do realize there are no easy answers here, but if you don’t have a really solid answer, and policy to back it up, normally routine questions can become problematic. Got any thoughts? I would love to both hear them, and use them to enlighten our readers.S|A
Note: It is SemiAccurate, not Semi Accurate. KTHXBYE.
Latest posts by Charlie Demerjian (see all)
- Microsoft gives Qualcomm’s Snapdragon WARTs - Dec 7, 2016
- Qualcomm announces 10nm Centriq 2400 ARM server SoC - Dec 7, 2016
- Broadcom fills out StrataDNX with three new chips - Nov 30, 2016
- Qualcomm’s Quick Charge 4.0 us based on USB-PD - Nov 28, 2016
- Intel outs lots of AI focused bits at AI day - Nov 17, 2016