throwawaysml

I suspected this might get downvoted and wanted to make clear I'm not demeaning the 90% of Google developers, but I failed, so I deserved the downvote. Just to make it clear I'm aware of where my comment failed to express what I was trying to communicate. Will try to be less lazy next time. It's hard to explain the purpose Google built Go for without considering where it's not used, and I failed to write a good comment.

True, and a fine line which is hard to judge from the seats we sit in. But I agree that a US company cannot be associated with the Kremlin or Beijing or Riad in 2017 without consequences. Maybe in 2023.

I just don't want to see a McCarthy witch hunt when a lot of the money floating in Wallstreet and the Valley is of dubious origin.

I would equally oppose taking money from one of the CIA VC firms, even that one where Dan Geer works. And I kinda have this idea that by now enough money is floating around valley from the money made in the valley that it should be possible to use that "pre-laundered" (if we may call it that) capital.

That said, I have no idea how a startup would vet the VC's capital backers, or what they should do when they find out some of it came from beijing, moscow, tel aviv, and you want to give it back to disassociate your startup.

Sure, but aren't we forgetting that Russia is like China in that over a certain size of business you are in no way able to operate unless you are in agreement or cozy (more business) with the Kremlin? This is in stark contrast to the US where it's no problem for a company to operate HQ in Oklahoma and loudly oppose the current DC government. I mean, if we assume Milner wasn't backed and planted by the Kremlin for a moment, then one could consider it impossible for Milner to be untied to Kremlin if based primarily in Moscow, no?

Thanks. Basically I want the sturdy laptop backpack you got from HP 10 years ago with a clever way to pack clothes and ideally waterproof bottom (rubber/plastic) so that I don't worry when having it sit on random floors at airports or train stations.

Having a subdivision of the main compartment bigger than 2 is also important. Laptop, papers, cloth, food maybe.

I'm in Germany and looking for a shop nearby where they have a diverse selection of bags to inspect. I'm kinda tired of ordering bags and having had to send them all back. Amazon is threatening a penalty for returning too often, so there's that :).

First world problems, right?

True and the more surprising aspect is that to land a develop position at Google you need to be on top of all CS theory and fresh in memory, just to unlearn everything and use Go for unspectacular business/enterprise projects, unless you're on certain teams like V8 or DeepMind for example. I think Go is meant to replace Google's Java coders with Go coders and have a language that fits exactly into the mold of their coding guides and rules for their monorepo and all the business/enterprise code written by the hordes of the rest of their developers.

Google's also opensourced Abseil, their C++ standard library (not meant to completely replact STL, to be clear), which contains all kinds of classes which were copied into many existing Google C++ projects in some form or fashion and sometimes incompatibly (e.g. Google's StringPiece found in several projects).

Either way I applaud them for doing a lot to open source projects. It's not something we can take for granted.

Is the difference to Chinese and Saudi money floating in the valley that we expect violent repercussions for anyone that opposes some aspect of the financiers' wishes? Or is it just that Russia is the most outspoken big nation that takes a different stance on critical issues where, say, Saudi and DC are in agreement. And that this is a danger to post-WW2 US influence? Genuinely curious.

Back when Milner invested in Facebook I viewed it as potential money laundering and didn't think about it too much.

If Java had good AOT and fast startup coupled with comparable initial GC heap sizes, it could have had a better chance at fighting off Go. Java will not go away and many teams that adopt Go also migrate to other languages at some point if their projects outgrow the capabilities of Go and the pain gets too strong.

I'm partial to GHC's language extension model over a cornucopia of Go pre- and post-processing tools like Java had (e.g. AspectJ and all the tools making use of annotations? or Go processors using comments). It will be easier to improve GHC's GC or complete OCaml's multicore branch than bring Go to the current century of proven programming language features. Go has found a niche as a replacement for C and Python in network programming, which is great. It just doesn't scale as well with project and team size. OCaml's multicore project also introduces algebraic effects (comprehensive alternative to monadic programming) to the mainstream, so I can't wait for OCaml multicore to land in mainline.

Past experience makes me wary to risk spending that much and be disappointed two trips later. Lifetime doesn't mean they give money back if unsatisfied, except one or two companies. Also, upping my budget didn't magically reveal viable options either. So I could have skipped the price tag mention and conclude that it's just hard to find the perfect bag.

I've seen the Go Ruck GR2 and it doesn't fit my requirements.

And it's made thinly veiled racist agenda mass acceptable as seen in this year's CSU (and to some extent CDU) communications. I'm scared a little, to be honest, though I continue to trust the checks and balances of the system supported by the rational minds of the greater population.

Am I alone in finding it unsurprising but unfortunate that all those addons to the official go toolchain are created by everyone to paper over the limitations of the Google Go implementation (which naturally reflects Google's development process and needs more than anything)?

EDIT: E.g moving .git/ back and forth or adding extra vetting/linting tools instead of extending `go vet`.

Go is becoming the new Java from what I can tell so it's good to know enough to find your way around and patch a Go project if needed. Being the new Java, despite being less expressive, explains the uptake in the enterprise and startups.

Graal, Truffle, SubstrateVM, Sulong

Is there a succinct explanation as to what those four projects are and how they would be combined in the hypothetical implementation you mention?

The idea is to replace all of the JDK with a construction of those four, right? Would it still need a traditional JVM to execute the resulting toolchain or replace it fully?

I'm not a bag geek, but I know firsthand how hard it is to find the right bag, if you care about functionality more than form.

For instance, I still haven't found the right carry on backpack that's sturdy, doesn't cost $200+, allows me to take clothes for a couple days and a laptop. It's either too big, too expensive, too heavy, too fragile, take your pick. So I can understand how someone can become a bag geek if they have to travel professionally on their own (without assistants who carry your baggage). Bonus points if I can detach a small laptop bag to take with me in order not to leave it at the hotel with the rest of the bag. So far I've always had to carry around the backpack and leave clothes in the hotel room. Pack, arrive, unpack, carry laptop, repack, go home.

It's ironic since there are more content creators now if we count all the users of the silos.

My theory is that it's all about convenience and network effects. So if a fully decentralized IPFS based/alike social network would come out with mobile apps and was marketed as the way to communicate, it could have success. The big fault such projects make is to stress the technical merits and forget to play the psych-advertising game.

I've seen non-technical people have the most number of messenger apps on their phones, and some even advocated the security of their favorite messenger. Technical folks seem to be harder to persuade to join a messenger network to communicate with a new person in their life.

So I want to say once we will have a couple more high profile data breaches and centralized SPOF downtimes, people will care about security and availability, though history doesn't support they would.

Didn't Google try to solve that with circles. I never used Facebook or any other "social network", so I'm probably unaware of some details, but I read about circles when Google+ came out and it sounded like a solution for exactly this problem.

I know, but how come we have now people from AfD in parliament in Berlin who openly gave interviews and public talks where they proclaimed the state and existing system has to be rebooted/abolished completely? Have they walked the thin line right at the edge that it was barely legal?

Do you think Verfassungsschutz (state dpt enlisted to protect the Constitution, and basically a spy/police agency) decided it's better to let them in for now and continue spying on them? Past events with informants and NSU do not fill me with confidence that Verfassungsschutz is doing their declared job.

It's clear why people vote for them, since they play the FUD strategy well, but I'm confused there hasn't been a concerted effort to expose them and their agenda.

Any insight?

tl;dr: good idea, but not applicable to many FOSS developers due to local laws.

Please keep in mind that the legal complexity of accepting donations prevents many of us from doing so. This is largely a result of local laws and how the IRS equivalent is set up. Take Germany for instance. The tax code in Germany discourages freelancing in favor of running a >10 person company or finding steady employment. The main reason is what taxes one pays for "real employees" and how that lands in the national support system accounts. This is a valid justification on the state's part, but a FOSS developer living in Germany who isn't already doing freelancing and has all the tax complexity and insecurity taken care of is hard pressed to start accepting random donations. The first problem you'll encounter is them declaring you a false freelancer if you have a "single customer receiving bills" or forcing you to deal with freelancing tax paperwork for accepting a couple hundred bucks a month.

Still it's a good idea and I'd suggest to take out the middleman for those micropayments. I don't understand what GitTip or Patreon add in value besides acting as the payment service. I mean, you won't be able to ask for money back, will you?

Hey, if I could accept donations without concern, I might buy hardware or expensive hosting services that would benefit the project, but most FOSS developers do it as a hobby paid out of their regular paycheck.

When I visited the R&D center of a German car manufacturer, we were checked at reception and needed special permission to carry the devices we had with us for the presentation we were there for. If the software lead (the host) hadn't cleared the way and taken responsibility, we would have needed to leave everything at reception like most visitors. We were only allowed to take in one laptop.

The surprise here is that the daughter was allowed to take her device into the development office. I suppose the father did the same as our host years ago by signing off and taking on liability for anything that might happen. In that sense, and I don't want to sound harsh, he unfortunately had a momentarily lapse of judgement and kinda failed his father and employee duties. Hard to balance them at the same time, but maybe certain Apple employees need to operate like, you know, Air Force R&D personnel. You go to work, but you never bring home or talk about work.

I know what you mean, but for this to be more widely used than Tor there is a need for me to know that nodes I operate only process stuff that won't be identified by 3rd party to sue me into a jail or financial ruin.

When I say whitelist, I do assume prefixes in the DHT (or whatever design they're using) can be used, as otherwise the whitelist may also grow too big to be practical. If however it will gain the concept of domains or such (which is also kinda a prefix), a whitelist will also be more practical.

Tor exit nodes are not operated by many in various places where they would/could because of concerns of the exit node addresses being enough to legally ruin operators' lives, even though technically it's a pure transit. Nobody sues the county because someone committed vehicular manslaughter on their public road. But because laws are skewed against the Internet right now, all someone needs is an exit node IP to make you regret.

If, and there's little to analyze/go by right now public, nodes and everyone is totally oblivious to what packets are transmitted; and if also the packets stored on nodes' disks are encrypted/sealed, then one could assume a filterless system to be practical.

This is all speculation, based on the little info there is. I really hope the team has come up with better designs that obliviate the concerns surrounding Tor, and ideally also not suffer from Freenet like slowness.

EDIT: Of course, once you have a filter, you will need to deal with the responsibility like Youtube does. If you do not know what's on disk or passing through, which is the ideal technically, then it would be best for the Internet and free communication. So, I'm not sure if a filter is a good idea, if that means you get subpoena'd and held liable for enabling one too many whitelist subscriptions.

For that purpose I would think optional whitelisting would make more sense. Instead of spyware lists you subscribe to in uBlock Origin, you would subscribe to a vetted, whitelist of known to be generally acceptable Orchid content.

Others might want to have empty filter list to be a complete transit/peer.

Therefore I don't think there's much room for users who want to blacklist but are not rather looking for a whitelist.

The blacklist will grow and grow, while the whitelist size will be pretty stable.

That said, I haven't used it, so the implemented blacklist approach might already support the above cases and be sufficient.

Yev, would you ever consider opening the software and hardware design of the vault?

It seems that you could lead this and give the industry a standard way, with vendors producing the fitting hardware, and allowing others to maintain their own vaults. Maybe this is a business advantage you don't want to lose right now. Then, I'm sorry to have asked :). If not then I think you might benefit by lower prices if the hardware is readily available and not a Backblaze custom production run.

There are many places who cannot for technical or won't for political reasons offload their storage to S3 and friends and need it all locally most of the time which would use a vault or two.

These things are very hard to find concrete data about for outsider, which I am one of, but at least nVidia has published that they're building their new Falcon controller chips based on a custom RISC-V architecture. It's hard to tell how widespread it is since they are not buying chips from Renesas or MIPS and looking at the board wouldn't necessarily show you there's a RISC-V chip in there. All we need now is a high performance variant of lowRISC to combine it with its safety features, port seL4 and MirageOS to it. Then would get finally get back what we had with the old Burroughs machines and enjoy safety and performance without having to choose.

Anyway, they announced this publicly in 2016 so I assume it's already hidden in some nVidia hardware. I also read about a storage vendor (Seagate? or WD?) working on a transition to RISC-V, but I cannot find it right now.

https://riscv.org/wp-content/uploads/2016/07/Tue1100_Nvidia_...

The reasons cited sound like those we heard about Linux and Samba or Linux and Apache in the past.

Hah, I never understood the appeal of AARCH64 either. It's not like it's some super clean ISA or that they won't run into similar power-vs-IPC conumdrums when they start to create x86 competitive server CPUs.

I've always been fascinated by the ARM and PPC embedded support on FreeBSD, which reflects the real market and not some fantasy land. Which also explains the sad state of the POWER port right now.

I'm fairly certain that RISC-V will eat ARM and it already started on the bottom end with embedded chips being replaced with RISC-V. nVidia, storage (SSD, HDD) vendors already have RISC-V silicon deeply embedded, hidden in sight. I feel like it will be the FOSS "revolution" all over and more progress because there's no need to pay stupendous amounts of licensing fees to ARM Inc. One could have done this with OpenSPARC too, but for political reasons or technical limitations of the ISA, it never took off.

My prediction is that many embedded ARM chips will be replaced by RISC-V. It won't be free because someone has to create the desired chip IP, unless you're happy with the free ones, but there will be competition that doesn't involve a parent licensing agency grabbing all the money for an ISA they invented 30 years ago. Good times to be a compiler codegen developer :).

Your other comment about NFLX makes you seem suspiciously and hilariously like a sock-puppet account so I can't actually be candid on that now.

I really wondered why they didn't mention POWER at all two weeks ago in the blog post. And now this post allowed me to raise the question. I'm curious what you think I represent for me to be a sock-puppet.

I cannot prove it, obviously, but this is just a throwaway account of a single person who doesn't represent anyone and definitely isn't paid to participate on Hacker News. Neither am I employed by any large or well-known organization to have some kind of intent. I'm just a programmer who wishes we had more POWER machines just like Raptor is trying to achieve.

I did feel uncomfortable asking the question as it might be obtrusive and thought I managed to word it defensively enough. Sorry, looks like I failed :).

For getting vendors to do things, it helps that I have a 8-figure HW spend.

Thanks for answering!

That's encouraging to hear, although sounds like I would need to be the operator of a video hosting or cdn network or HPC cluster before I could request a consultation with engineering executives. I'm most excited you have good relations with Chelsio since I love their NICs and FreeBSD drivers. I couldn't care less about Intel NICs tbh.

Speaking of NUMA, I'm glad that AMD Threadrypper will force more developers to get familiar with it, since it's the first mainstream single-socket system that uses it. It will bring it to the attention of more developers than just those who tested regularly on multi-socket machines of their employers.

That's cool, may I ask how you persuaded them to bring the FreeBSD port up to date?

BTW, the new POWER chips can run in a new endiannes mode and therefore lift the big-endian requirement, if that's a concern of any code base.

I don't know, I mean, programmers spend so much money and little value hardware just for brand value and such, so the money is there. But those that spend that amount of money on computers are seldom in the group of those who have an interested in a POWER machine.

There used to be remote access offered by IBM and Uni of Oregon for open source developers to do porting work.

When the recent Netflix FreeBSD kernel optimization article was posted, all I could think of was why they didn't use a POWER machine for its higher memory throughput. Same would be true for SPARC, but that hardware is even less available and I wouldn't trust Oracle with my money or engineering effort.

Then I checked to see FreeBSD's POWER port and it didn't look as readily supported as Linux is, so I assumed that's one reason. Contrary to SPARC, POWER servers can be had from more than the main vendor and it's the only high performance server where you have a chance to get the source code for the microcode and firmware in auxiliary mainboard chips, leaving just the GPU and NIC. This alone I would think would be a reason for safety-conscious users to consider and buy it instead of trying to disable Intel ME.

I still think Netflix ought to evaluate POWER9 (POWER10) and possibly improve the FreeBSD port, and would love more industry support so that little teams like mine stand a chance to get reasonably priced POWER9+ servers.

The completely free aspect is one reason I believe Kollab is using it for their servers.