31–36 of 36 posts
I wish Rust people (like Steve here) would stop saying that there's no new research in Rust. It's just totally 100% not true. Large parts of the Rust type system involve awesome new ideas. Additionally, as I said once to dherman, language design is _always_ research -- putting things together in a new way is basically all research is.
I find it interesting that he contrasts free software with open source as "ethical position vs. cost reduction", apparently without ever considering cost reduction as an ethical position. (Enabling people to do things for less cost seems (to me), to be at the heart of any non-zero-sum strategy for improving the world. Otherwise you're just re-shuffling the same cards that have already been dealt.) I also thought it w…
> apparently not considering the value he gets from other people without paying them.
Oh, I think about this _all_ the time, trust me.
I wish Rust people (like Steve here) would stop saying that there's no new research in Rust. It's just totally 100% not true. Large parts of the Rust type system involve awesome new ideas. Additionally, as I said once to dherman, language design is _always_ research -- putting things together in a new way is basically all research is.
As far as I know, the newest bit is regions, taken from Cyclone, in 2001. Am I missing something else?
But my more significant point is the second one, which is that novel language design is research, period.
Earlier quoted context omitted.
As far as I know, the newest bit is regions, taken from Cyclone, in 2001. Am I missing something else?
Regions as a general concept go back at least to FX-87 (in 1987 :). But the way the Rust type system works is quite different from Cyclone, for example. Cyclone, for example, treats memory in the heap quite differently than Rust, and doesn't have borrowing (at least not obviously). But my more significant point is the second one, which is that novel language design is research, period.
I am not anti-capitalist, so I am sure Steve and I would disagree quite profoundly on many things. However, one thing we both do agree on is the distinction between free and open source. I do not particularly care for open source, as it is a nice way to provide unpaid work for other people to exploit and make lots of of money off of without providing compensation of any form in return. That seems extremely unethical…
If you've read some of the alternative literature on capitalism you might think that profit is by definition providing "unpaid work for other people to exploit and make lots of of money off of". It's a rather clever inversion of private property rights that exploits the medium of commodity exchange - to pay someone a fixed wage for their commodity called labour time and exploit their produce to turn a profit. Open-so…
Oh, other remark - the ethical perversion that MIT/Apache/et al allowed is that billions of dollars are made off those licenses, providing very excellent lifestyles to selected people, and there is no compensation ipso facto to the actual authors outside of (maybe) recognition in license.txt files hidden away on phones. Recognition does not (as a first order effect) help bills get paid (or a better life for author (those might be second order effects)), nor does it help the world in general (except for the product itself).
Consider the effects of open source on the economy (positive), and the fact that thousands of people do this for free, enabling companies to essentially get free labor (and then have the gall to want unpaid labor proof in interviews).
Stallman approaches this topic grounded as a question of ethics and morals, with an imperative to share information.
I approach the topic in three ways: "Am I taking this thought-stuff and making fat cash and not giving anything back to the person who made it?", and "Can the end user repair (or find a repair person) the product I have provided them?", and "If this is a good project, can it be maintained and recovered in thirty years time". I don't think anything but the GPL (and friends) gives that position.
You can also say to my position that it's terribly unethical to ask people in each company that wants to do X to do the same project (but duplicated for each company), rather than having an open source project that has liberal licensing and having the engineers share work. Wasting other people's lives is a harsh thing as well. :-)
I wish Rust people (like Steve here) would stop saying that there's no new research in Rust. It's just totally 100% not true. Large parts of the Rust type system involve awesome new ideas. Additionally, as I said once to dherman, language design is _always_ research -- putting things together in a new way is basically all research is.