Magic Leap – Separating Magic and Reality
kguttag.com
kguttag.com
That isn't illegal if he isn't lying or anything, but whether he is right or not, he isn't an unbiased source.
He was also eviscerated here and never replied, though he replies to almost every other reddit comment:
https://www.reddit.com/r/magicleap/comments/5c337v/magic_lea...
(disclaimer: I own a small amount of EMagin from before his post)
I'm not sure what you think eviscerated me. My information is factual.
I try to answer all responses but I am getting a LOT responses and I may have missed some at reddit. It is best to go to the comment section on my site.
It turns out that Magic Leap does use Mico-OLED in one of its prototype, but they are also using DLPs and LCOS. My subsequent posts point this out. I warned people that it was just a hunch. I kept my stake in eMagin at least for the time being, but as I said it was a small stake for me.
BTW, my website is down right now. I found out that there was a lot of traffic from Hacker News while trying to see what was going on which is what brought me here. I'm on hold with tech support right now.
After he showed you the caption on the images all you came back with was a statement (via an edit to your post) that the first image was still an "artists concept" (it isn't, highlighting the projector's field of view and pointing out that it was done in the caption isn't anything like an "artist's concept").
And then for the second image you said it "might be real."
That means you are essentially accusing the researchers and saying they "might" be faking the second image and committing research fraud. Their caption says, "the bottom photo is unaltered."
(shameless edit: I have VR and mixed-reality experience including relighting the real-world with game-engine lighting. If any Magic Leap guys are browsing the thread and looking for third party content devs or recruiting, send an email to mljobcandidate@gmail.com!)
OK, I was answering a lot of questions that day and I went to the direct links the person gave to the jpeg images and not the paper. The first one was obviously altered but the picture I saw did not have the caption. Same for the second, I never saw the caption or that paper before I went back to check tonight.
As for speckle, it would not have shown up in the low resolution pictures. It might also not be there due to the vibrating fiber (some types of vibration will essentially eliminate speckle but usually it has to be the screen that is vibrate).
BTW, my website may be down for a while.
You're of course welcome to comment here, but HN commenters are required to remain civil and substantive, and especially to refrain from personal attacks.
It's a fairly factual and benign set of comments to my eyes.
No one ever is
What I've learned - I think - is that the magic leap video demos use a totally different technology than that which they hope to bring to market -- seems pretty analogous to Theranos to me.
Here's the kind of thing he has been writing on reddit:
>I'm driven by facts and evidence and not by some big name person investing money. These people sometimes invest based on ego (which pretty much explains Google Glass). I would suggest you watch the movie "The Big Short" (not totally factual but it gets the point across). [1]
My eyes rolled when I read that quote from him: he started his whole debunking-Magic-Leap blog series by making a disclaimer that he was buying eMagin and trying to pump them up by showing that Magic Leap might be using them.
Only egotists would invest in hyped-up Google glass! That's why he wishes he did:
> The last time I played this game of “what’s inside” I was the first to identify that a Himax LCOS panel was inside Google Glass which resulted in their market cap going up almost $100M in a couple of hours.
A little bit of ego there, huh? Why, he was personally responsible for $100M of market cap and wishes he could have bought in so that he could dump the position during the hype, knowing that Google Glass itself wouldn't take off in the long run.
And even after missing out on that opportunity, he missed another:
>About the time Himax stock hit its high, I found out about the Seeking Alpha article and posted a comment reiterating what my blog post had stated that I didn’t think the Himax panel was used in the newest prototypes (there were also comments from others stating the same). Himax then stock dropped back about 16% or about $135M in market cap. (Gad, I should have shorted before I wrote that comment ).
None of this is some huge moral indictment; just be aware the guy is now openly thinking along these lines when he writes technical analysis on his blog.
[1] https://www.reddit.com/r/magicleap/comments/5fyq6b/some_more...
That said, I have serious doubts about their ability to secure the hardware and software from intruders who believe it is just another objective reality in which they may speak and act for others.
Whether or not different teams are building the marketing website (something you assert knowledge of but apparently don't actually have; companies are constructed all sorts of ways), I'd hope it's true that the core engineers have looked at the company's website. So either they didn't notice a problem, didn't care, or cared but didn't have enough power to get things fixed.
Now it could be that the Magic Leap project ends up amazing anyhow. Maybe the engineers noticed and cared, and so are good engineers, but the marketing side is indeed separate and is kind of a shitshow. But hopefully not enough of one that it impedes a good launch.
But I don't think it's unreasonable to evaluate a company by the materials the put forward specifically for evaluating them.
Nonsense. How is a developer going to notice this? It's a simple misconfiguration of an HTTP header. That's it. Their AR and hardware developers may have never even done web work and may not know how HTML 5 video even works to even THINK this could be an issue.
I work for a company but I don't check their main, consumer facing websites much if at all because that's not what I work on. So you're telling me if there is an issue with one of them it reflects poorly on myself?
> But I don't think it's unreasonable to evaluate a company by the materials the put forward specifically for evaluating them.
Since when does HTML help you evaluate AR and hardware engineering?
Get off your high horse.
It may be just a HTTP header issue, but someone's not on the ball. I'd be annoyed if my mobile bandwidth was getting eaten up by a obvious problem. But credibility judgements are always dependent on what the viewer deems to be relevant criteria. So both of you are correct. A simple problem with the website says something about the company, not the tech. It is a minor thing but perhaps means something.
One company I worked for had a website created by a 3rd party, a cookie cutter thing, which I wasn't too impressed by given my experience in websites. It also had some little glitch that I found and pointed out on my first day. And that turned out to be a difficult year and a half... some personal stuff on my side, but the company was also managed in a substandard way. Lots of people were frustrated, it turned out. A tech company with good front end dev/designers ought to have a spiffy website.
I'd be inclined to give ML most of a break, but a tech company that has problems with its marketing side, especially when marketing something like ML is so crucial... what percentage of their initial customers is going to be people in tech? Gamers? Who do notice these sorts of details?
When you view in a mobile context it does a download of much smaller videos and only once. It works differently for mobile and requires you to trigger playback.
> A simple problem with the website says something about the company, not the tech. It is a minor thing but perhaps means something.
So email them and maybe they'll fix it? Unless you'd rather continue using it as a yard stick of dubious value.
> I'd be inclined to give ML most of a break, but a tech company that has problems with its marketing side, especially when marketing something like ML is so crucial... what percentage of their initial customers is going to be people in tech? Gamers? Who do notice these sorts of details?
You're putting way too much stock into such trivial things. Unless you're specifically looking for an issue you're not going to find this. The amount of people who would find an issue like this are on the edge of edge cases.
You seem to have an axe to grind against the company.
No, I said I'd be inclined to give ML a break. I don't have an opinion about ML other than I'm tired of hearing about them. I'm not the parent commenter.
The main point I was trying to make was that credibility is assessed by people in different ways depending on their expertise / knowledge base / etc. So someone knowledgeable about websites would notice such a thing and the average person wouldn't. So arguing over whether something is an absolute signifier of credibility or not is sort of pointless.
> Get off your high horse.
Sorry, but I'm not interested in your drama.