So, let's break down the fields of CS for which this should be applicable:

* Systems: This won't work except for the few systems projects that are entirely in-RAM AND will work on tinycore's kernel version

* ML: This, I can see, especially with the seeming focus on dataset management. Much ML is compute-bound and the overhead of using the FUSE FS's is hopefully negligible.

So, is this focused on ML and ML-using code and experiments? If so, I think that should be clarified. I think a lot of systems folk will be (rightly or wrongly) turned away from it due to the seeming overhead of the various hyper* extensions. Not to mention that they are all written in Node/JS (Again, rightly or wrongly, many systems folk will not want to run their stuff on platforms written in JS)

I like the direction this project can go, but there seems to be a lack of focus or direction in your mission right now.

> So, is this focused on ML and ML-using code and experiments?

Your completely missing the point. Please look into 'Computational Science' (or Scientific Computing, or Numerical Analysis), that applies to 80%+ of all disciplines that exist today (e.g., computational physics, comp. biology, comp. economics, comp. aspects of engineering disciplines, the list goes on).

However, this still isn't clear in their website. I will give them the benefit of the doubt since they are early in their project, but I think it would behoove them to nail down their mission sooner rather than later.

This is probably what I get being in the CS bubble. =)

But I agree, it would help if they expand on this from the pure CS point of view. Especially if they mention things like containers, CS people would be interested in finding out what they're up to.

Seriously though, this kind platform is a critical component in scientific reproducibility. The dream is that we can have code, data, and the results of the composition of the two in the same revision control system. A minimal layer to allow the execution of linux software would support the use of legacy code and binaries in this new platform. Javascript has its advantages, but it's a waste to build a data RCS and require all functions on the data to be written in it.

And to go a bit further, it's not just for science. For example, you could write a HN clone in dat. I could fork it and get both your code and all the posts.

Microwave works on line-of-sight and weather patterns easily mess it up and drop the link.

That's why you usually also run your microwave data over a fiber link as backup. Of course, when microwave works, it gives you a decent improvement in latency, at the cost of bandwidth.

Also, I found that the best way to get rid of feedback was to get an ear mold rather than using an "open-fit" mold. This is a clear separation between the speaker and the microphone and pretty much solves the problem in my experience.

EDIT: I noticed late that you addressed the cosmetic issue in your post. I don't see the elderly changing but our generation just might.

I'm youngish (mid 30's) and recently had my hearing aids replaced and realized I had a strong preference to stick with fairly visible behind-the-ear units rather than something more "discrete". I want them to be visible so that people I'm interacting with will be more sympathetic about repeating themselves and may make an (possibly unconscious) effort to speak more clearly.

The line of thinking that got me over being self conscious was "Lot's of people walk around with assistive devices for their vision...why should I be embarrassed about the same thing for my hearing?"

Also, I work in a noisy environment where hearing protection is mandatory, and I find I have less trouble understanding people if I can see their face.

What a good point.

I have an elderly friend who has suffered complete hearing loss in one ear after an infection and the other ear can only detect very very low frequencies, and he's constantly saying "PARDON?". It must be very difficult to hear ANYTHING going on, other than the rumble of lorries and buses. I wonder if they could put a pitch-shifting circuit in his hearing aid to shift sounds up/down so that they fall within his hearing range, whilst not shifting frequencies already in that range. That would help significantly, surely?

Just thinking out loud.

If you read the article, you'll see that's more or less what most modern hearing aids do, via a technique called multi-band compression.

Edit: Actually, here anigbrowl, an audio engineer, states that this is not how multi-band compression works: https://news.ycombinator.com/item?id=8854142

Assuming he is correct, my above statement may well be wrong.

There is no pitch shifting in multiband compression - pitch shifting involves moving the frequency up or down by a number of cents, semitones, octaves etc. It's the effect used to get the "chipmonk" voice (high-pitch and squeaky) where a normal voice is fed into a pitch shifter and it is shifted up or down. It is also how harmonisers work, where they work out the frequency you're singing at and shift it up 7 notes (or an arbitrary amount) so you can sing and get a harmony of yourself.

I've had hearing aids like that for almost a year, and "fairly visible" is a stretch; they're pretty blasé and don't stand out.

This is especially true of modern "Receiver In The Ear" (RITE) models where instead of a tube carrying sound, you have a very thin wire going into your ear canal.

It's probably worth distinguishing between two kinds of phenomena that might be described as carrying a stigma:

- Something might lead other people to mock or otherwise denigrate you for exhibiting it. Being fat is a good example here; fat people get a lot of messaging from society that they're worse people for being fat.

- Something might carry no real significance to the rest of society while still being viewed, by the individual, as painfully embarrassing. There's a traditional view that women don't like to wear glasses because they think the glasses ruin their looks. I don't know how well that currently corresponds to reality; I've known one girl who really hated her glasses for that reason and another who, not needing glasses of her own, liked to take other people's and wear them -- but that's the prototype of a "category two" stigma: a woman who hates wearing her glasses even though no one around her sees anything wrong with them.

I suspect that hearing aids are firmly within the second category, which means getting people to wear them "openly" should be doable.

You suspect wrongly. Having seen the attitudes to my father change when he wore one. Ranging from outright verbal abuse, to assumptions of stupidity & senility.

(I have no idea what is the correct term for "behind the ears", I hope it's understandable.)

Off-topic: While I was there, they asked employees to submit ideas for a new hearing aid marketing slogan, with the incentive of a free vacation to Vegas going to the person who submitted the one they used. For some reason, I did not win the vacation with my suggestion: "Stick It In Your Ear!"

At the low end when eating out in restaurants occasionally having wait staff ignore him and asking other folk at the table "what would he like", or people assuming that he couldn't hear and talking about him — to at the high end having a guy shouting "deaf fuck" at him repeatedly on the street for no obvious reason.

I'm not trying to say that this happened every day — especially the outright insults. But it was enough to be noticeable.

I suspect, as @hibbelig commented, age had something to do with it.

Ben Heckendorn is known for very clever console modifications and has forayed into accessible controllers in recent years. He sells a custom made one-handed controller for a few hundred dollars, and this may be worth a look!

Also, I find lxappearance an easy and quick way to set GTK themes via GUI rather than having to edit the .gtkrc files for every option.

Funnily enough, this problem has been solved before! If you look at distributed file systems, especially something like Coda (http://en.wikipedia.org/wiki/Coda_(file_system)), they are designed to make the local computer a "thin client for storage". Basically, local storage is used as a cache for the main copy on the server, and this behavior is transparent to user applications via the FS driver.

Dropbox uses a user-space program to sync the files and cannot intercept system calls. As such, they had no choice but to synchronize all of it rather than bringing things in "on-demand" and releasing local copies that are not used. Nowadays, this can easily be done via FUSE driver in Linux. I do not know if something like FUSE exists on other platforms though.

Overall, I personally love the idea of a local machine being a "cache" for the server copy. However, the technical challenges are greater than a simple mirroring scheme, and there may be UX issues as well.

It looks like there might be something called OSXFuse which might be more recent and maintained, but I haven't looked at it, it also might not be.

I know Intel has software for their NIC's to enable userpace networking, but I don't think any HFT uses it. It may be all right for hobbyist experimentation. I left the industry about a year ago, so my information is a bit out-of-date, but my former employer used either Solarflare cards with the OpenOnload stack (very good cards and awesome software) or the Mellanox CX series with VMA stack (amazing hardware with mediocre software. VMA is now open-source, so perhaps that situation has improved).

Note that these cards are a few thousand dollars each, so out-of-reach for hobbyists.

For the latter (which matters in routers, for instance), netmap runs on everything (either natively or with some emulation). The intel 10g cards (which DPDK of course supports) are around 3-400$ i think, the 1G cards start around a few 10's of dollars (not that you need userspace networking at 1G).

Alice/Java => "101"-type class

C/Perl/Shell => Intro to Systems

C => rest of the systems branch

Java => Data Structures

Standard ML => functional and logic classes.

Many of the more esoteric electives used their own thing. For example, I used Serpent in Computer Music Systems.

The curriculum changed in my final year. The new students now do:

Python => "101"-type classes and basic data structures and algorithms

Standard ML => parallel data structures and algorithms and other functional and logic classes

C0 transitioning to C => Systems courses.

I do think our distributed systems class was experimenting with Go at some point.

With the quashing of the slow metadata performance (http://lwn.net/Articles/476263/), XFS seems to be all-round just as good as ext4 but with more future-proofiness for large volumes. Keep in mind that RHEL releases are supported for around a decade.

Looks like XFS supports concurrency better (both on the request side, and on the kernel<->backing store side).