amirmc

> "No-one expects companies to pay the full whack of tax."

Why? I don't understand why this is considered acceptable. I'm expected to pay the full whack of tax (income, VAT, etc), but somehow it's taken as given that corporations will pay 'less than they're supposed to'. How can we fix anything if this is the prevailing attitude?

I read a fascinating book on tax havens and their pernicious influence on global affairs. I thoroughly recommend it to anyone who wants to understand more about how such systems work -- especially those who don't think tax havens are an issue.

http://treasureislands.org/

In that case, why would anyone remove money from a corporate entity? All those who could afford it would just act through corporate avatars. I think this already happens to some degree.

I didn't mean to imply that the claims were not strong. Just that they're more careful/specific (which is a good thing for this kind of work).

Also, saying a piece of software is "unhackable" is akin to saying a ship is "unsinkable".

They also have http://sel4.com -- why wouldn't they use the .com as the canonical site instead of a more obscure .systems? (I say this because sel4.systems is linked from within the original article).

Also, the link http://sel4.systems/FAQ/proof.pml is a 404 (linked within the article where it says "Last year, Heiser’s team proved mathematically that their kernel is unhackable."). I guess this is meant to link to http://sel4.systems/Info/FAQ/proof.pml (where the claims are far more measured).

To be fair to the article, it should be obvious that it's about SoundCloud by the end of the first paragraph (& thus the 'we' refers to 'us at SoundClound').

I do agree that the title here is out of context but it seems any modification of titles is seen as editorialising.

Folks interested in this might like to know that ContainerCon also had a session on Containers and Unikernels. http://sched.co/3YUJ

A write up and audio from that session is also available.

http://thenewstack.io/the-comparison-and-context-of-unikerne...

Previous thread at: https://news.ycombinator.com/item?id=10115013

There's also been some discussion about this on the MirageOS mailing list [1].

There are definitely operational things to work on (e.g. more convenient deployment to AWS, amongst other things). One of the best ways to get feedback on these is from the people -- like the OP -- who try things out.

End to end toolchains make a huge difference but taking the time to come up with the correct opinionated tooling is important. Being opinionated and wrong could torpedo the uptake of such tech.

[1] http://lists.xenproject.org/archives/html/mirageos-devel/201...

> "And, facebook, imo, is more of "get shit done", rather than "break shit". This was more of the latter."

He didn't 'break' anything.

> "Look at all of the other social networks that tried to solve this in the past and failed."

FB has a network effect, which a a major reason all those products/services get used. However, it's not really fair to imply that FB is 'succeeding' at all the things you listed. If it were, then there'd be little talk of existential threat from things like Instagram/Whatsapp etc...

> "Can't comment on the content as I was too annoyed with the UX to read past a few slides"

Then just don't comment at all.

Someone generously put a slide deck of their talk online. A deck that was actually designed to be delivered by the author, probably while standing next to them, at an event. Obviously, it's going to be slightly out of context for people who come across it on the internet (esp with no video/audio to accompany it).

If people wrote their slides for the internet first, imagine how boring all the conference talks would be.

That reasoning could also be applied to anyone who's suffering in an abusive relationship (whether emotional or physical). It would be inappropriate then, just as it's inappropriate here.

In addition, not everything has to be decided solely by where one spends their money. It's entirely possible to put pressure on an organisation through other means.

Most people do not have a background in security so this is anything but a 'duh' moment for them. The more news like this that makes it to the mainstream media, the more hopeful I am that regular folks will learn the state of security today.

> "I don't know if the editor changed after 2004, but me and many other readers have the feeling that after 2004-5 the quality dropped incrementally."

I'd concur with this. Even though I disagreed with some of the articles I found them to be well written/edited and I'd learn about different viewpoints (so I subscribed). They even published a style guide. I don't recall when but I started to feel that they (ironically) stopped following their own style guide and around the same time the quality began to drop.

There's a nice post about entropy in such (virtualised) environments, with specific regard to unikernels.

https://mirage.io/blog/mirage-entropy

> "One can't make a self-driving car by edict, ..."

An edict can put a man on the moon.

http://history.nasa.gov/moondec.html

My point was to refute the assertion that it's "because developers don't run companies in general." If we're going to add lots of caveats to what 'developer' means, then we're essentially making a 'No true scotsman argument' -- (i.e. Zuck hasn't worked anywhere else, and is ostensibly a manager now).

True, but there's a difference between 1. your employer having a bunch of information about you and 2. your employer signing up to a ToS with third parties that effectively hands over all your info to them.

Point 2 is probably not unusual but were starting to see (imho) more pernicious attitudes. Examples include the current Windows discussion, where your employer may be perfectly happy to sign (on the employees behalf) that your data be sent off to Microsoft (and used for whatever Microsoft deems 'necessary'). There was also the announcement of Facebook at Work, which I'm sure would be more than happy to cross-correlate the 'work' you with the 'personal' you, in order to 'provide a better service'.

In both these scenarios, I'm sure the company will act to protect itself from exposure, but it's unclear what choices the employees really have.

That comment is just a way of saying 'they' are not like 'me' (but trying to make it sound general).

Facebook, led by a (former) dev, has gone all-in on open plan offices - which is a direct counterpoint. That people at the top call the shots is tautological.

http://www.forbes.com/sites/kevinkruse/2012/08/25/facebook-u...

Network effects mean that there are few pragmatic options left. The GP may well be free to use alternatives but at what cost? Should they be selecting future employers based on the employer's choice of software?

I may well be free to go live in a (digital) cave but presenting it as a viable choice merely exacerbates the existing problem.

> "... meaning that a customer that sits half a day with a laptop costs Starbucks money."

If you consider that the largest cost is probably rent, then people sitting around probably add only a tiny amount more. It's only an issue if the shop happens to be be full and there are other customers who wanted a seat.

That person sitting there is probably doing Starbucks a favour. People are more inclined to go somewhere where there are other people.

That's an utterly ridiculous criterion/comparison. Of course, I'd turn around and go back home for the device that people expect to reach me on (irrespective of how 'smart' that device is - it's my phone after all).

The thing that smartwatches bring that fitness trackers didn't is a multi-function device. Each feature is incremental and by itself might not be compelling enough to care about long term, but add enough of them together and maybe then it can jump the chasm into everyday use (e.g. telling the time + fitness tracking + alerts + payments + quick comms + ???).

> "Frankly, I live my life in front of a computer on a desk. My calendar and a million other things are one click away and in a format far more convenient than a phone ..."

This exact argument also applied to smartphones, yet we've seen what can happen if you put a computer in people's pockets and let people figure things out.

I'm not suggesting the watch will displace the phone (the way the phone can displace laptops for many users) but I believe there's huge scope for augmenting it.

> "Yes, they are selling bunches of them, but I fail to see this as a long term product unless they pull a rabbit out of a hat."

Selling bunches of them is enough to indicate that something will happen in this space (i.e. people are already spending their money on it). Whether this iteration or the next will be the one that nails it remains to be seen. For example, the upcoming changes to allow native watch apps (as opposed to only having the UI there) should allow developers a bit more freedom.

> "There are certainly a lot of karma coming to a lot of people who deserve it, ..."

That's an extremely judgmental comment.

I also added a feed of your OCaml posts to OCaml Planet [1, 2] (just so that doesn't catch you by surprise) :)

[1] http://ocaml.org/community/planet/

[2] https://github.com/ocaml/ocaml.org/commit/8df5da9b6fe1ba4c43...

The very important difference is that at a job you get paid regardless. You can even shift your mental model such that "it's just a job" and start leaving the crap 'at work'.

In a startup (esp if it's your own), the psychological and emotional stakes are significantly higher. Also, there's no reason that politics can't find their way into a startup you're running. You just might not notice it because you've positioned yourself at the top of the tree of monkeys.

Have you come across Rumpkernel? I'd recommend looking at that since it can work with legacy code. There's a static-website tutorial that describes how to use Nginx as a unikernel.

http://rumpkernel.org/

https://github.com/rumpkernel/wiki/wiki/Tutorial%3A-Serve-a-...

I wrote up a post about the event with some additional information, which now has a few clarifications/errata at the end. I'd welcome any feedback!

http://amirchaudhry.com/unikernels-polyconf-2015/

> "...to benefit the US"

More specifically, I think it benefits US corporations. There's a sleight of hand in politicians/lobbyists presenting it as benefiting the US as a whole.