So I joined Google in 2006 (December) and one of the things I had pitched to a few folks before joining was something I was calling "JaDE" which was short for "Java Application Development Environment". Nearly everyone I pitched it to thought it was just a 'thin client' ala Citrix's offering, and Brad Templeton called it a 'diskless workstation' as he knew my history with Sun and Sun's flirtation with that sort of thing in the early 90's.
I didn't think it anything like that at the time, thinking instead of it as a sort of appliance like client rather than diskless (it had local flash storage), it was basically what smartphones became, but with a keyboard and display[1]. When Android was being developed and a number of people in the platforms area were getting excited about it, I talked with the Android folks about forking it into something along the ideas of my JaDE concept. They weren't very receptive, they had a phone to build. I did the usual stuff, created a p page [2], put some stuff in the wiki etc. Also saw Jeff Nelson's previous work (which had gone nowhere), read the politics around the whole Android effort at the time and abandoned my efforts to push it any further.
Other projects inside of Google with that level of detail did get pushed further by people interested in moving the ball down the road. When one of the platform folks left to join this "secret" group which turned out to be ChromeOS I was amused because yet another group of people had figured out what, by then, was a pretty obvious idea. Since I happened to have a compatible ASUS Eeepc at the time I downloaded one of their early builds and played around with it. When they got some traction inside the company the Android group went ape shit (as expected) and that led to some interesting changes.
But lets come back to Jeff for a moment. He was at Google for 2 years according to his LinkedIn profile, which means he joined, got slotted, and left. (a common path sadly). But he is also currently the CTO of 'stealth startup' which means he is out pitching to investors with his co-founder(s). Its so much better to pitch as the 'inventor of ChromeOS' than it is as '2 years at Google, then consulting for 4 years.'
I clearly remember he had a 'p' page for his project (I found it when I was thinking about my version of the idea) so he may be assuming that someone read it and said "Hey this is a pretty good idea, we should build this." And started building it. Since it was never a good idea at Google to give anyone else credit for your ideas perhaps Jeff is making some unwarranted assumptions. I could easily see him believing whole heartedly that he invented it, just like I can believe nobody in the ChromeOS group has ever heard of him.
Google is just that kind of place.
UPDATE: see the patent claim in a sibling comment, that is priceless!
[1] Yet another case of being too early!
[2] Internal database of "projects" being worked on.
Fascinating story, Chuck, thanks for the insights.
One dumb question: can you translate "got slotted"? I'm guessing it has to do what team you end up in, but I have a feeling there may be a more nuanced meaning to it?
Sorry, Google tried to solve the problem of 'mismatched' titles in their own unique way. Since "Senior Engineer" at some persons previous job might not compare to what Google called a "Senior Engineer", Google delayed determining what your "pay grade" was until after you had done some work.
So Google would hire you, you would spend 6 months being useless trying to figure out what the heck was going on, then you'd get on a starter project or some first "real" project, and 12 to 18 months later your manager would put you in front of a committee of folks who would look at your work output, and then pick what "slot" you were.
If you were slotted "lower" than you had been hired at, they had a talk with you (and it meant you couldn't get a pay raise until you had been promoted twice). It also had an annoying tendency to slot people who had been "Principal Engineer" or "Distinguished Engineer" at a previous company into "Junior Engineer" at Google, which really pissed them off (and sometimes lead to "management action" in the form of a formalized "performance improvement plan").
Basically the more senior you were, the more likely you were to wash out. Since Jeff was a "Principle Engineer" at eBay prior to Google I'm guessing that this was his path as well. Probably got slotted as 'senior engineer' (which is three levels below Google's equivalent of Principal Engineer) and said "Screw this, I've paid my dues already, see ya!" But that is just speculation on my part.
I have to admit that I love stories like this as well as the Tesla refutation of that NYT review.
I think my interest in the nature and impact of our reliance upon human testaments first developed when I became engaged in theological discussions.
Here we are with all our instant access to information, our pictures, our videos, our blogs from the man on the street, our DNA samples, etc., and we can't figure out if one guy got legitimately stranded by his electric car or if another guy created a new Operating System -- yet you can go to church on Sunday and listen to lots of people who will swear all day long that a woman was impregnated by the sole deity of the universe two thousand years ago. This knowledge is based upon a translation of a story written many years after the supposed event took place.
My impression from what I've read is that Jeff did work on a linux fork which was known as Google OS internally (I think there were even public news about google working informally on a "Google OS" in 2006-07), but that never really catched on and was never turned into an actual product. Years later, the idea of a web-first Chrome OS appears, and either a) someone reused Jeff's work to build the core and nobody paid much attention to it or b) both things have absolutely no connection and Jeff is mistaken in assuming his work was turned into Chrome OS.
Antoine Labour: I am one of the 3 original engineers on the project. The other 2 have left the company, so I will claim, if you allow, that I'm the most senior engineer currently working on the project.
I have never heard of this guy.
He has certainly not given any input on the project, to the extent of my knowledge. For that matter, the Chrome OS project itself, the one that ended up growing and actually shipped products, didn't exist in 2006 when he claims he invented it.
Reading his blog post, it sounds to me that what he's really claiming he put together bits and pieces of a linux distribution based on the concept of running off of a ram disk. This has pretty much nothing to do with Chrome OS, I don't see in what way his project could have morphed into what Chrome OS is today.
"I have a copy of a May 18, 2007 (not April) email from Jeff announcing his project on a wide-distribution list. It does not have the subject line "Google OS"; it's actually called "Guppy needs testers". Given that it begins, "I'd like to publish my Linux distro, called Guppy, for some testers to try out and provide feedback. Guppy is a little side project to build a Google distro of PuppyLinux for USB.", I'm inclined to believe it's the first such wide-distribution email. The phrase "Google OS" is never mentioned; neither are "browser" or "Chrome" or "Firefox" or "webapps". The closest we get is the single sentence two thirds of the way through that "I set up some desktop shortcuts to Google Apps", which is literally the only time the email mentions anything close to Jeff's claim above that "all of the functionality came from webapps; performing any operation on the desktop launched a Chrome window to one of many webapps." In fact, the email explicitly mentions that you can install OpenOffice if you want, which seems to contradict this portrayal a bit."
Google is a big place. There can be multiple people working on similar ideas (see Chrome OS and Android). If the projects are small it would be common for them to not even know about each other, at least for a little while.
If you tried to enforce that everyone who wanted to work on similar things had to work together you would burn a lot of energy arguing about "similar" and relatively minor differences in direction.
Sometimes it is easier and better to just let there be separate efforts and the market (both the internal attention market, and in extreme cases, the external actual market) can decide.
The news of Google OS around that time was almost certainly conflated with Goobuntu, Google's custom Ubuntu LTS install. The assumption was Google was prepping its own Linux release, but there was never any hint it would be made public.
If a single person at Google understood/reviewed the 'Google OS', and then later also authorized/reviewed the 'Chrome OS', then the strong form of your (b) option, "absolutely no connection", is impossible.
Even without explicit sharing of code/people/documents, the earlier project would have served as an investigatory prototype and proof-of-concept, that indirectly encouraged or dissuaded certain later directions. So these 'true father' disputes get into shades of meaning and clashing-but-valid perspectives pretty quick.
At some point you need to draw a logical line, lest you arrive at Linus Torvalds inventing ChromeOS since he wrote the kernel for it way back in the early 1990s.
I have no knowledge of ChromeOS history other than what has been published here, but from my reading of these things it does seem pretty disingenuous for Jeff Nelson to claim to have 'invented the Chromebook'.
The wording "invented the Chromebook" is aggressive, especially for collaborative projects that collect many contributions and work on shared themes. But, journalistic accounts and promotional bioblurbs typically dumb things down, in exactly this way.
And, if at any point Google asserts that Nelson's '662 patent covers the Chromebook/ChromeOS or similar products from other companies, then it's a defensible statement, at the level of fuzzyness that's used in these contexts. It should be interpreted as, "invented [key enabling technology for] the Chromebook"... like when people talk about who "invented the iPod" or "invented AdSense".
Speaking as a googler from that era:
It's much more likely Jeff did something independently, and it never ever got reused elsewhere.
In fact, it's very likely nobody working on chrome/chromeos ever knew about it.
We've yet to find anyone associated with Chrome OS who even knew Jeff Nelson existed. Perhaps there is someone, but there are few people left who could possibly fit that profile. Who's more senior than the original developers, Sundar Pichai?
The indirect relationships I posit doesn't even require anyone to remember Nelson's name, just for the chain-of-organizational-learning to exist.
I don't know anything about the Google org chart. But when Google OS was a project -- and Google financed the Nelson patent application -- who was he describing his work to? Did any of those people, or even people they passed ideas to, later help make the decision to prioritize/fund Chrome OS? If so, that's more than "absolutely no connection".
Did Brin/Page/Schmidt take interest in and see presentations about Nelson's 'Google OS'? And then later on 'Chrome OS'? That would also be more than "absolutely no connection".
Even if the 'Google OS' project was a mess and the main lesson that percolated down from it was, "don't do it this way", it could have been an important precedent... even if Nelson's name is forgotten. That "there are few people left" who might have personal recollections cuts both ways.
Note that these are all general arguments about evaluating why people's impressions differ, and how in founding disputes everyone may have a reasonable basis for their claims based on their own limited perspectives. I don't have specific knowledge of Nelson or these Google projects, and on the internet, exaggerations, hoaxes, and identity-games are always a possibility.
Google+ doesn't have permalinks for comments, otherwise I would have linked to Antoine Labour's comment towards the bottom. Says he is the most senior engineer still working on Chrome OS and has never heard of Jeff Nelson.
The title of this submission: "Chrome devs deny Jeff Nelson's role in the creation of Chrome OS"
Quotes from the actual article:
"I'm somewhat skeptical of this."
"It would be nice if some Chrome OS folks more knowledgeable than I could shed light on this, because I truly don't know."
Analysis: The linked article has one developer (not "Chrome devs") confessing that he simply doesn't know what the facts are (not "den[ies] Jeff Nelson's role in the creation of Chrome OS")
Conclusion: Don't write your own submission headline, you will get it wrong.
Yeah, I don't think that's very prudent of them. This happened seven years ago, and as they say, many of the involved people left. What if one of them did have a meeting with this guy?
Well, for starters, one of the guys who commented has been there since the beginning.
As for the rest, this is very unlikely. Think about this for a second. They'd already started a project called ChromeOS, and they then decide to have a meeting with this guy about the idea of a web browsing laptop (or whatever)?
That doesn't make sense.
I have my own recollections of when/how the ChromeOS project started (since they asked for legal and open source advice ), and it pretty much matches other folks in that thread. I never heard from this guy, and I had heard from now-executives (they were just regular folks back then :P) about the idea/plan.
So, while theoretically possible still, it's pretty unlikely.
I expect this is simply a case where a guy thinks what he did got used elsewhere, and it wasn't.
There were a lot of independently developed projects that did kinda-the-same thing back in 2006-2007.
There's some sort of patent involved - http://www.google.com/patents/US8239662 - "Network based operating system across devices", filed in '09, issued in '12, assigned to Google.
Oh that is sweet! Google was finally waking up to the patent reality after 2007 and encouraging folks to file patents on their projects. (I've got one filed for a laser pen pointer toting telepresence robot somewhere out there) It looks like Jeff took his ideas and sent them into the patent guys who dutifully filed them and voila, "invention" of ChromeOS without having written a single line of code of it :-)
Reading through the claims it does look like ChromeOS uses those concepts so I guess Jeff gets the last laugh after all.
I tend to agree that Google doesn't file enough patents.
For such an innovative company, with hundreds of different products in several industries, and many of the best and brightest employees anywhere, their patent portfolio is miserable.
Regarding, "It looks like Jeff took his ideas and sent them into the patent guys who dutifully filed them and voila"
Well, for one, this wasn't just "ideas". I created an operating system and used it myself for over a year, pitched it to management, sent it to a company wide email list, and got head count assigned to the project.
Aside from that, Google has been very contemplative about what patents to file or not file.
Without going into any great detail, I wrote 2 patents on my Google OS operating system - and only the one was ever perfected.
I wrote 3 or 4 other patents on unrelated projects that were all filed as provisional patents and allowed to expired.
At least 1 of the expired provisional patents was very, very commercially valuable, in my opinion, but Google apparently did not recognize that.
Pathologically sure, I'd say. He attempted to change the Wikipedia article for Chrome OS to include his name, and has been spreading stuff all over the Web saying that he created Chrome OS. In addition, his "blog" only includes one page, and that page only describes his relation to Chrome OS. Also, if he really did work on the project, why would the senior chief engineer have no idea about him?
It could just be a joke, for all we know. People screw up Wikipedia for humor as well.
My reasoning is that he wants to add a big achievement to his credentials and he's using the Chrome OS fiasco as leverage. He's trying everything to get recognized, even illegitimate means (such as defacing Wikipedia).
Why he posted that on the Star Wars page, I haven't the slightest clue. But it's safe to say he does want to get recognized. Some people are egoistical (speaking hypothetically here) and even if they've failed at something, they want to get recognized for it regardless of others who may even have contributed exponentially more to a project. It's a form of psychological catharsis for failure. It's happened to me before.
> My reasoning is that he wants to add a big achievement to his credentials and he's using the Chrome OS fiasco as leverage. He's trying everything to get recognized, even illegitimate means (such as defacing Wikipedia).
The Star Wars article change was a joke - that was corrected about 5 seconds later. I'm part of the generation of Star Wars fanboys. I apologize if that offended anyone.
Right, here's the summary he has on his profile verbatim then:
"Mr. Nelson invented Google Chrome OS while working at Google in 2006. He has 19 years of experience in cloud computing and Big Data analytics. He has written two books and many magazine articles on Java and cloud computing. He has extensive experience in search, SEO, and eCommerce industries, building highly scalable web services, and leading world class engineering teams at such companies as Google and eBay. He holds a Masters Degree in Applied Mathematics."
MTS is the name that was given to folks who hadn't been slotted yet.
So either he got slotted and didn't like his title enough to list it, or he never got slotted (which would be very odd for 2 years).
Linkedin is pretty confounding in that regard, because usually you can see at least a public profile of a linkedin member if you're not logged in. Usually I open an incognito window in Chrome with Apple+shit+N and read the profile there. You could also open a 2nd browser where you're not logged into linkedin. Nonetheless, I hate it when sites try to base features around the technical shortsightedness of their customers.
I am not surprised some people are calling this claim into doubt since its was work done back in 2006 and 2007. It sounds like many of these questions are being raised by people outside of Google or individuals who joined the project after 2007.
First, let me reassure you that while I was not able to walk out of Google with my email history or design documents, Google sent me long correspondence about the Chrome OS patent after I left Google. I don't believe this correspondence is covered by any NDA since it was sent to me after I left Google, and I have retained all of that correspondence.
I published the work called "Google OS" on a company wide email list in April 2007, with subject line "Google OS", and received hundreds of positive comments from other engineers at that time. I also held a tech talk on "Google OS" in May 2007. Anyone at the company in April 2007 will likely retain a copy of the original "Google OS" email, search for subject line "Google OS" in April 2007.
The operating system that I invented, as described in the April 2007 email on a company wide email list, was a webapp-centric chopped-down Linux with a Chrome browser front-end. The operating system had almost no applications installed on it, instead all of the functionality came from webapps; performing any operation on the desktop launched a Chrome window to one of many webapps. The original April 2007 version of the operating system that was published on a company wide email list was substantially identical to the publicly released Chromebook product, as was the writeup in the original 2007 email.
Further, back in 2007, I had meetings with Jeff Huber (VP of Google's consumer products group), Larry Page (now CEO), several other directors and managers, and even presented a techtalk - all of these meetings on Chrome OS project. By the end of 2007, I was working with a product manager, and together we were able to convince management to launch the Chrome OS project and assign head count to the project by the end of 2007. In August 2007, my product manager and I even met with an external hardware vendor to have exploratory talks regarding their interest in distribution a Chrome OS laptop.
All of this is verifiable both from the email record, such as the April 2007 "Google OS" email to a company wide mailing list, and also by those who met with me on Chrome OS in 2007, including Larry Page himself. Further, I retain the hard copies of the correspondence with Google that was sent to me after I left the company on the subject as well, which I don't believe is covered by any sort of NDA.
I hope this clears up some of the controversy. Again, I am not surprised that many of you who either worked on Chromebook or joined Google after 2007 have never heard of me. Once I left Google, there would have been no reason to continuously bring up my name as the original inventor, and I have chosen not to come forward until the patent was finalized and published. I apologize to those of you who may have been confused by this.
So to summarize:
1. I wrote the original Chrome OS and published it on a company wide email list in April 2007.
2. I wrote the patent in 2007, which is now accepted and published by the USPTO as of August 2012 and shows my name as inventor.
3. I convinced management to launch and assign head count to the original Chrome OS project by the end of 2007.
Would you mind posting (potentially redacted) copies of those emails, just so that we can get an idea of what was involved in GoogleOS? It would be nice if some of the people you worked with piped up to tell us more about this too. I suppose that they will see what is happening and do so some time soon anyway as this is getting quite a bit of exposure.
This seems legitimate to me and, frankly, the responses from Peter Kasting and other Google employees look to be immature and childish. Perhaps most Google employees are not as professional and high caliber as they appear on the surface.
There were jokes made in that thread, but can you point to any of Peter Kasting's actual comments that would qualify as immature and childish? I only know Peter's work on chrome/webkit bug reports and occasional sites where "pkasting" shows up (he has a single comment as an hn user), but I've only known him to be communicative and supportive, and this is more than a small overreaction to paint him like this when all his post amounts to is "this doesn't sound right. anyone know more?" and his additional comments are specific and not emotionally charged. Hopefully the irony in your post isn't lost on you.
As for the patent, it doesn't seem all that chromeos specific, actually. For instance, the claims are very focused on an OS that is or can be stored server-side, with the client fetching an image from the server (that includes the user's preferences?) on startup and dumping it again at shutdown. The patent process tends to make ideas more generic as they are recorded, but even a generous interpretation of this doesn't really cover chromeos or the OS he described in the blog post (at least the part he most emphasized, the idea of keeping everything in RAM). ChromeOS doesn't get OS updates that differently than other operating systems (and certainly not as new images every time it connects to the update server), and the preferences sync system stands totally separate from the rest. This doesn't invalidate the blog post, but I don't really think the patent backs up the specific claim of "inventing chromebook".
They seem like rather abrupt and well-time for something revisions.
I also haven't come across any Chrome OS engineers, past or present, who even knew of your existence prior to this controversy. How do you explain this?
A strong defense which I found convincing, especially because the effort required to disprove it is now more significant than "Any of you guys ever hear of this Jeff Nelson? Me neither." At the time (with zero information) I thought that on balance he was likely to be inflating his claims.
Also, because I was annoyed at all the unfunny, low-effort jokes posted in the comments on the linked G+ page, which felt like piling-on, I confess that I'm pleased at the turnabout. Good twist.
It's not too surprising to discover that the original engineer behind a product is not known to later generations who build upon, sustain, or otherwise refactor the original invention. Indeed, unless the previous engineer sticks around, it's almost never the case that future generations even know their names, unless they happen to have their fingerprints on the current SCM (which may or may not be related to the original project).
What mailing list?
I have email from April 2007, and I don't see it.
I do see other emails from you, but mostly about random stuff. Nothing I can find that appears like you describe.
(I'm not claim it's not there, i'm just trying to find it :P)
Peter Kasting tracked down the original company-wide email. It was actually from May 2007 and titled "Guppy needs testers" - for those of you working for Google who would like to look it up. Since I left Google in January 2008 and had to turn over all of my records, I'm working from memory here.
The entire experience of having work that I did 6 years ago questioned, has been interesting.
However, the invention, the communications, the meetings, and the OS itself are all very well documented both inside and outside Google. Further, I'm fairly confident in the integrity of Google management.
I hope I have responded in an entirely professional manner, and I'd like to assure everyone that I hold no grudges and, particularly for those Googlers who weren't aware of my work, I'd welcome the opportunity to buy you a beer sometime and fill you in on everything that happened in 2007.
You cite emails which do not appear to match your claims. You cite technology which does not match your claims. Is there another person you worked on this with who you could ask to comment on your work -- former manager? product manager? any peer engineers? So far, you've called exclusively on upper management-types who traditionally cannot comment.
Also, what specifics do you feel like you invented pertaining to Chrome OS? Just planting the see of a web-centric Linux distro (on a USB stick...)? Or are you claiming all its boot, graphics, security, and OS-level innovations?
Hi guesty, I apologize I didn't remember the subject line of an email I sent 6 years ago.
If you are a member of Google, you can almost certainly track down a copy and run the actual operating system.
I had correspondence with hundreds of Google engineers at that time. Obviously if I had access to my email record, I could provide the contact details of all of those engineers, but that was all turned over to Google when I left the company.
Its not evasive. I'm stating if you are at Google, you can go look at the actual operating system itself. Also, copies of the slide deck we presented to management - on many occasions - is also almost certain still sitting on the Google file system. Its not like anything ever gets deleted at Google.
I too created ChromeOS :-) Let me explain;
So I joined Google in 2006 (December) and one of the things I had pitched to a few folks before joining was something I was calling "JaDE" which was short for "Java Application Development Environment". Nearly everyone I pitched it to thought it was just a 'thin client' ala Citrix's offering, and Brad Templeton called it a 'diskless workstation' as he knew my history with Sun and Sun's flirtation with that sort of thing in the early 90's.
I didn't think it anything like that at the time, thinking instead of it as a sort of appliance like client rather than diskless (it had local flash storage), it was basically what smartphones became, but with a keyboard and display[1]. When Android was being developed and a number of people in the platforms area were getting excited about it, I talked with the Android folks about forking it into something along the ideas of my JaDE concept. They weren't very receptive, they had a phone to build. I did the usual stuff, created a p page [2], put some stuff in the wiki etc. Also saw Jeff Nelson's previous work (which had gone nowhere), read the politics around the whole Android effort at the time and abandoned my efforts to push it any further.
Other projects inside of Google with that level of detail did get pushed further by people interested in moving the ball down the road. When one of the platform folks left to join this "secret" group which turned out to be ChromeOS I was amused because yet another group of people had figured out what, by then, was a pretty obvious idea. Since I happened to have a compatible ASUS Eeepc at the time I downloaded one of their early builds and played around with it. When they got some traction inside the company the Android group went ape shit (as expected) and that led to some interesting changes.
But lets come back to Jeff for a moment. He was at Google for 2 years according to his LinkedIn profile, which means he joined, got slotted, and left. (a common path sadly). But he is also currently the CTO of 'stealth startup' which means he is out pitching to investors with his co-founder(s). Its so much better to pitch as the 'inventor of ChromeOS' than it is as '2 years at Google, then consulting for 4 years.'
I clearly remember he had a 'p' page for his project (I found it when I was thinking about my version of the idea) so he may be assuming that someone read it and said "Hey this is a pretty good idea, we should build this." And started building it. Since it was never a good idea at Google to give anyone else credit for your ideas perhaps Jeff is making some unwarranted assumptions. I could easily see him believing whole heartedly that he invented it, just like I can believe nobody in the ChromeOS group has ever heard of him.
Google is just that kind of place.
UPDATE: see the patent claim in a sibling comment, that is priceless!
[1] Yet another case of being too early!
[2] Internal database of "projects" being worked on.