HNHacker News
TopNewBestAskShowJobs

catwell

1,287 karma · joined March 5, 2011

Software @ Finegrain.ai. Before: inch.fr (acquired by Septeo), eFounders (Hexa), Lima, Moodstocks (acquired by Google).

https://catwell.info

submissionscomments
catwell··on A Visa for founders, engineers and investors willing to join France
Off the top of my head, large-ish French tech software companies (I would not call them startups at this point) include: Criteo, Dailymotion, Leboncoin, Withings, Parrot, BlaBlaCar, Sigfox... You could also include older companies like Murex or Ubisoft, as well as OVH. You may also want to look at other fields where France is recognized for its expertise (energy, pharmaceuticals...).
catwell··on A Visa for founders, engineers and investors willing to join France
The truth is it is a lot more complicated than you make it sound.

There are some schools which are relatively young and focus on coding. Historically they were expensive (for French schools), even though one of them is now free thanks to the generosity of a French billionaire. They recruit students with different profiles than the traditional "Grandes Écoles" and they get much less recognition from French employers, especially large companies. Because of that, their top students may indeed be inclined to move abroad, where degrees matter less.

In the more generalist, traditional engineering Grandes Écoles, however, the phenomenon is a lot less prevalent. Some students do move abroad, but my experience is that it is not correlated to their competence. Those with propensity to found companies are more likely to go to the US, but that is mostly because the ecosystem there is like no other. On the other hand, you may have noticed that some large US companies (Google, Facebook...) have been opening research labs in Paris, while top French companies such as Criteo send their marketing department to the US but keep development in France. This is not only because talent is cheaper (others have already explained how low salaries for not mean low costs) but also because top technical talent doesn't necessarily want to emigrate. This is cliché, but Fabrice Bellard is still living in France, and Yann Le Cun moved back. :)

catwell··on Just how smart is an octopus?
Also, they are part of the few edible maritime species whose populations are increasing (http://www.humanosphere.org/environment/2016/05/rapidly-chan...).
catwell··on Dd is not a disk writing tool (2015)
One of those Unix tools that would deserve to be better known is dcfldd (http://dcfldd.sourceforge.net/). It is basically dd with extra powers including on-the-fly hashing, progression, multiple outputs...
catwell··on Nvidia and Audi aim to bring a self-driving AI car to market by 2020
Yes. This is why https://news.ycombinator.com/item?id=13111516 was on the front page last month.
catwell··on Lua use with HN users?
In addition to that, I think those should be mentioned:

- Software Defined Networking. We are starting to see LuaJIT in particular being used to write very high performance network code in userspace (e.g. Snabb).

- IoT. Lua has always had a niche as an embedded language, especially on weird small hardware platforms. It is known for being extremely easy to port and using few resources. It is what you reach for when you want slightly more than what a Forth gives you. See for instance the first Tessel, NodeMCU, the Sevenhugs SDK...

catwell··on Organ donation becomes opt-out in France
Yes, I added a source (the first one) which confirms that. 32.5 % of the time the relatives said no, but polls showed that it should only have been 21 %.
catwell··on Organ donation becomes opt-out in France
You had to consult relatives and ask them: "do you know if the deceased was opposed to this"? This meant the way to opt out was to tell your relatives. It has been the case for 40 years. [1]

There were posters in most hospital waiting rooms telling people to do this (which I found creepy as I found myself there a few months ago, waiting for news on whether a relative would live or die, by the way). See [2] for an example.

Now the only way to opt out is to send snail mail (!) to be added to a nationwide file. It should become possible to do it over the Internet someday.

Fun fact - we have a pin card that has been designed to store that information since 2006, but the feature has never been made accessible to the public. [3]

[1] (fr) http://www.lemonde.fr/les-decodeurs/article/2016/12/22/don-d...

[2] (fr) http://fr.slideshare.net/amfe/guide-du-don-dorganes-2015-de-...

[3] (fr) https://www.agence-biomedecine.fr/Carte-Vitale-2-et-don-d-or...

catwell··on Ask HN: What do you want to learn in 2017?
Economics. I have started to get interested in them in 2016. Like most topics I get into, I started by looking at the history of the field. I also read Basic Economics. Now I would like to start taking actual lessons, either Coursera or maybe even physical classes at some point, if I can find good ones compatible with my work schedule in Paris.

I also want to get up to date with the most important advances in my field (distributed systems and algorithms). I did follow research in 2015, but not in 2016, so I have to catch up a bit.

catwell··on Ask HN: Books you read in 2016?
I finished Keynes, Hayek: The Clash that Defined Modern Economics (Nicholas Wapshott), started in 2015.

I read:

* Turn the Ship Around!: A True Story of Turning Followers into Leaders (David Marquet)

* Joy at Work: A Revolutionary Approach To Fun on the Job (Dennis Bakke)

* Ne vous résignez pas ! (Bruno Le Maire - French politician)

* Dealers of Lightning: Xerox PARC and the Dawn of the Computer Age (Michael Hiltzik)

* Disrupted: My Misadventure in the Start-Up Bubble (Dan Lyons)

* Making Things Happen: Mastering Project Management (Scott Berkun)

* Basic Economics: A Common Sense Guide to the Economy (Thomas Sowell)

* The Success of Open Source (Steve Weber)

* Weapons of Math Destruction: How Big Data Increases Inequality and Threatens Democracy (Cathy O'Neil)

* Programming in Lua (fourth edition - I read every edition)

I started reading (and will probably finish by the end of the year) Overcomplicated: Technology at the Limits of Comprehension (Samuel Arbesman).

As for what I recommend, it depends what you are into, but I would say I really enjoyed Making Things Happen, which is a must if you have any kind of project management to do, and Basic Economics.

catwell··on Skip Lists Done Right
Yes, that is correct. The code checks the encoding at runtime ( https://github.com/antirez/redis/blob/8444b46d20ef9c8de3f7e2... ) and converts between the structures as needed (see calls to `zsetConvert`). The alternative (for large zsets) is called zskiplist, the manipulation code is directly in t_zset.c and the structure is defined at https://github.com/antirez/redis/blob/8444b46d20ef9c8de3f7e2...
catwell··on Skip Lists Done Right
Data structures in Redis have evolved a bit in the last 7 years. One of the main workhorses is indeed something called a "ziplist". The implementation (especially the format description at the top) is worth a read: https://github.com/antirez/redis/blob/90a6f7fc98df849a9890ab...

For those who are familiar with pre-2.6 Redis code, there used to be something called zipmap, but it was deprecated in favor of ziplist.

catwell··on Elon Musk, Travis Kalanick, Indra Nooyi to Join Presidential Advisory Council
I didn't know that one, I did use http://twister.net.co/ some time ago (and gave up because of the chicken/egg effect).
catwell··on GRiSP – Erlang-based IoT hardware platform
With 16k RAM, you won't get much. But the ESP8266 has 64k + 96k RAM and runs things like Lua and MicroPython. So now imagine what you can do with the ESP32, which has 416KB RAM.
catwell··on DynASM
The backward compatibility issue Mike is ranting about in this post is not in the language, it is in the source code of Lua itself (actually a define in its configuration header file).

There was a very good reason for this change, which is that Lua introduced an integer type instead of using floating point numbers for everything. This means that LUA_NUMBER_DOUBLE and LUA_NUMBER_FLOAT do not really make sense anymore (when you check those you may want to check the type of the integer numbers or the type of the floating-point numbers).

In any case, porting luabitop doesn't really matter that much: if LuaJIT supported Lua 5.3 people would use the native 64 bit bitwise operators it introduced instead.

catwell··on Uber Now Tracks Passengers’ Locations Even After They’re Dropped Off
This kind of fraud happened to me three days ago. I filed a complaint with Uber and they resolved the issue. I certainly think something should be done in the product to help with this, but there may be better options, including letting the user click a button to declare that the trip is over.
catwell··on Lua: A Guide for Redis Users
There is no reason why modules would be easier to test than scripts. It is probably the opposite.

Modules are great when you want to use Redis as a server for something it does not support, but I would still use scripting for most things.

catwell··on How close are we to solving vision?
I am not sure this can be called "adversarial examples" but we sure know how to make the human brain fail at vision tasks, with optical illusions and things like the famous Invisible Gorilla test.

https://en.wikipedia.org/wiki/Inattentional_blindness#Invisi...

catwell··on How close are we to solving vision?
This is the same kind of question as: "are computers better than humans at Math." They are obviously better at some things related to scale: they can compute things much faster and derive solutions to equations much more easily. But the issue is: they don't really "understand" what they do. And that is why computers are still not better than humans at discovering things independently, even though a lot of proofs are now machine-assisted.

Similarly, machines are becoming better than us at recognizing some specific instances within categories of objects because they can know more of them, i.e. they have larger "databases". But they are still bad at learning new concepts on their own, even though there has been much progress on that front in recent years.

In general, I don't think it is a good idea to consider limitations in Computer Vision as "vision" issues; instead we should consider them as wider AI issues. Basically, ask ourselves: "could a blind Human solve this problem with the information that our current vision algorithms have"?

I wrote a more detailed response along those lines over three years ago at Quora, when I was still working in CV and the field was achieving its switch to neural nets and deep learning. I still think it is mostly relevant today.

https://www.quora.com/What-are-the-major-open-problems-in-co...

catwell··on Ask HN: Any good resources for finding software jobs abroad?
Networking.

Seriously, ask your social network contacts in the countries that interest you. Demand for good developers is high everywhere, odds are there are openings at their companies.

Don't just go and browse job offers on random websites, that's inefficient and a lot of good offers are not there anyway. Another option would be to subscribe to something like https://www.talent.io/. Or, if you know some, just get in touch with headhunters in the country. They often have very good offers nobody knows about.

catwell··on Homebrew 1.1.0
If you want Unix-like software, look at msys2 (http://msys2.github.io/). Its package manager is Pacman (the same as Arch Linux).

When I have to use a Windows machine I basically do everything in a msys2 terminal.

catwell··on Why does software development take so long?
This tweet was really spot on: https://twitter.com/jtolds/status/758716505320075266
catwell··on Kodak Ektra
But how much market share do they have among people likely to buy expensive digital photography equipment?
catwell··on The power of prediction markets
I don't have much insider information, but mostly the founders gave up on it and moved on to other things as their monetization plans did not succeed.

To monetize I know they had tried a model of sponsored predictions: http://blog.beansight.com/post/6171212673/beansight-hit-its-...

They finally ended up Open Sourcing the code: http://blog.beansight.com/post/85724048548/beansight-is-now-...

catwell··on The power of prediction markets
Pretty similar yes. I didn't know that one, thank you.
catwell··on The power of prediction markets
The "guess the number of jelly beans in a jar" game gave its name to one of the startups whose failure disappointed me the most, Beansight. In 2010, it was asking people their opinion on various topics, attributing them experience points when they guessed right, and using this information about the expertise of its users to improve its predictions.
catwell··on Final – A credit card built for the 21st century
I have been using something like this for years (a French service from my banking group called Virtualis). I have no idea why it's not more frequent.

The main thing to be careful about is not to use this to pay for things where you will be required to provide physical proof of card ownership later. For instance, some French railroad tickets can be withdrawn from machines where you must insert the card that has been used to purchase them.

catwell··on Mozilla has stopped all commercial development on Firefox OS
I used WebOS, and I agree. I really enjoyed my Pre3. I could never bring myself to like Firefox OS.

The best now is Jolla, which I don't think has failed at all yet (I have a Jolla C, I wrote about it shortly here: https://blog.separateconcerns.com/2016-09-16-thoughts-1.html).

catwell··on WoSign and StartCom: Mozilla’s proposed conclusion
There was a good start of an answer to that in Perspectives [1] (and later Moxie's Convergence), but sadly those projects are technically dead now.

I still think those were good idea, but nothing is ever going to replace CAs without the support of major players such as OS and browser vendors.

[1] https://perspectives-project.org/

catwell··on My love-hate relationship with LuaJIT (2015)
I have to agree here, however this is mostly not an issue if you use luacheck, assertions and unit tests.

Still, stronger typing would help, and this is why several people are working on adding gradual typing to Lua. The main initiative is probably https://github.com/andremm/typedlua

← PreviousPage 6 of 12Next →