Game_Ender

Do you have some press links?

Thought experiment: If you had such a system and it was good enough that people would not notice it’s use, would you advertise that? I would look carefully at the wording of these companies. Such a system would let you paper over all sorts of issues and claim zero “driver interventions”...

Has she consider using Uber of Lyft as they are right now?

UndoDB supports threading [0]:

"Handles concurrency

Record and replay multi-threaded programs, and those that use shared memory and asynchronous I/O."

[0] - https://undo.io/technology/features/

Phabricator has the wiki, code review, and ticketing systems tightly integrated. It's a clean, fast, and free alternative to JIRA.

https://www.phacility.com

You are assuming just because it's Pittsburgh pay from companies like Google, Apple, Amazon and Uber will not be competitive. If there is any discount on salary it's on the 10-15% range, not 50%.

Can you describe the build process? Building 100's of millions of lines of C++ in 20 minutes is quite a feat even today. How much parallelism and build distribution was used?

There is always the existential "Does anything really matter, everybody dies in the end?". This could have direct application, or as with many math concepts, in 50-100 years we could find an application for it. Despite all of that it doesn't have to matter, the process of exploration can have value in and of itself.

They are so liquid and predictable they are essentially just more salary. So a startup needs to not just have good equity, but great equity.

It's like getting a bonus for living somewhere more desirable.

I feel like S1 filings are always brutally honest like this. I believe it's a legal thing, they are afraid of future suits.

For Velodynes you get some spurious points that look like noise cuased by reflections or rain/snow. Any mature system already has some kind of filter so it's tractable. The rate is on the orders of 10's to 100's per second which is low compared to the million/second of the sensor.

Quanergy never shipped their product because they were not able to hit the required for performance for it to be usable. So far it looks like only Waymo and Velodyne have shipped LIDAR sensors good enough to the primary sensor on a self driving car. Everyone else is still at the "vaporware" stage.

WebGL has really high draw call overhead compared to native OpenGL. The rough rule of thumb is that you can do 2000% [0] the draw calls in a native app per frame than you can do in WebGL. It's not about execution performance it's about overhead of leaving JS and converting data to the format the GPU needs.

0 - http://floooh.github.io/2016/08/13/webgl-next.html

I agree that Differential, the Phabricator code review application, does not handle patch sets with hundreds of files or many thousands of lines of well. They don't optimize for that case much because they believe in smaller changes [0].

I think the local patch problem you are experiencing is due to the way the author submitted their diff. Normally if you want to locally try code you run "arc patch D123" and it will create a branch off that revision and apply the patch for you. If you look at this review [1] you see it has a link to the repository "Repository rP Phabricator", and in the "Revision Update History" and "Local Commits" sections it has direct links to the base and parent commits. If the submitter based their diff off non-published commits this won't work.

You can also reduce the need to directly pull down code by integrating a CI tool, code coverage information, and screenshots into your review process. All of those Phabricator supports.

[0] https://secure.phabricator.com/book/phabflavor/article/recom...

[1] https://secure.phabricator.com/D15358

Your comment hit me like a bolt of lightning. Every time I use a Freestyle machine I can't believe someone delivered a product that is so agonizingly slow. It's such a basic idea, why does it take what feels like 1-2 seconds to respond to Every. Single. Click.

Does this include ArrayFire Pro [1], mentioned on your doc pages? I am having a hard time finding the source to the pro versions on your GitHub page.

1 - http://www.arrayfire.com/docs/arrayfirepro.htm

I would love to see first class support for GUI and 3D accelerated applications with docker. Right now it's not easy to get it setup, and it's not entirely clear to me what the various performance overhead is of different solutions.

On MRAP front the military is getting rid of them for a good reason. They ordered the initial batch very quickly and ended up several different platforms [0]. None of them fit the bill perfectly and having several variants made logistics and vehicle operations difficult. They then did a real procurement program and ended up purchasing several thousand M-ATVs. The surplus MRAPs went into this law enforcement give away system [1].

0 - http://en.wikipedia.org/wiki/MRAP#MRAP_deployment

1 - http://en.wikipedia.org/wiki/MRAP#Post-war_reductions

Not adding much value here, but it is a lot worse. You get 5+ implementations everything, and every project is different just because.

A similar company is SeaBotix [0] they have a similar [1] thruster for sale but it uses a brushed motor. They have a newer generation product (not separately listed) that uses brushless motors that has integrated motor controller so all you need to do is send it power and I2C speed commands.

For people interested in this I recommend checking out the RobotSub [2] competition. Each team releases papers allowing you to see how the whole vehicle fits together.

0 - http://www.seabotix.com/

1 - http://www.seabotix.com/products/auv_thrusters.htm

2 - http://www.auvsifoundation.org/foundation/competitions/robos...

It does 3D scene and camera path reconstructions then re-renders the scene from different perspectives. It's not just "picking the best frames". The technical explanation video goes into the details: https://www.youtube.com/watch?v=sA4Za3Hv6ng

Took a little bit of reading but the OP is founder of company [0] making a massively scalable geospatial database. This is the closest article [1] I can find which has anything close to details. It lists the following technologies:

* A low level stack that bypasses the OS for at least disk IO (supposed to provide a 2-3x throughput increase)

* "polymorphic" space-filling curves for distributed parallel indexing

* hyper-dimensional spatial sieves [2]

* Allen’s Interval Algebra to parallelise SQL statements

More discussion on some of the OP's previous hacker news posts [3].

[0] - http://spacecurve.com/

[1] - http://www.it-director.com/blogs/Bloor_IM_Blog/2014/7/how-do...

[2] - http://www.google.com/patents/US20090182837

[3] - https://news.ycombinator.com/item?id=7482151

It's similar to the sizer concept in wxWidgets:

http://wiki.wxwidgets.org/WxSizer

Python doesn't have a STW garbage collector, it uses deterministic reference counting. So while slower than go you won't have random pauses in your server processes due to memory management.

Much of the current public outreach really does focus on people taking their eyes off the road and missing things [0]. Usually in pretty graphic ways [1]. The other main focus is navigation in the car, which pretty much everyone does now, so making that safer is a win.

[0] - http://www.ibtimes.com/texting-drivers-take-eyes-road-5-seco...

[1] - https://www.youtube.com/watch?v=Ss021L0hWU4

Is this openSUSE only, or can you build packages for other distributions with this service? It's hard to tell from their documentation or website.

EDIT: The openSUSE instance [0] says:

The openSUSE Build Service is the public instance of the Open Build Service (OBS) used for development of the openSUSE distribution and to offer packages from same source for Fedora, Debian, Ubuntu, SUSE Linux Enterprise and other distributions..

[0] - https://build.opensuse.org/

So it appears they do support different distributions, although I can't seem to find download links or examples.

Is there a specific API document for ycmd? I tried looking through the code base and did not see anything explicit. For emacs support the implementation will have to be in elisp, so having a python independent way of describing the API would be helpful.

As an emacs C++ user YouCompleteMe is the only VIM plugin that makes me jealous. I want the code navigation and completion you get with VS or XCode but within emacs. I know there are other plugins that try to do this, but as the author mentioned, all the repeated works leads to an overall mediocre results in most cases.

Maybe the front end of something like irony-mode [0] can be adapted to use the YouCompleteMe backend.

[0] - https://github.com/Sarcasm/irony-mode

You need better in app instructions. A user should not have to watch a video in order to figure out how to complete or save a scan.