jsolson

38 ms·

391.
by jsolson 11y ago
I tend to partition them into windows once the favicons go away. Otherwise mostly based on the ordering (which I adjust regularly). Much of the time they are effectively a queue of things to review/skim/familiarize myself with whi
392.
by jsolson 11y ago
I see this critique now and then, and it always makes me wonder what is wrong with your Chrome installation. I typically have somewhere between 200 and 400 tabs open, and I work on a wimpy little 11" MacBook Air. The vast majority of t
393.
by jsolson 11y ago
> The ability to reflash the phone and avoid triggering the unlock erase was a pretty obvious workaround. Or just replace one of the off-chip AHB/AXI devices and inject code via DMA. It won't work beyond the iPhone 5C, but I ha
394.
by jsolson 11y ago
Sure, there's an extremely well known vulnerability in the iPhone 5C hardware. Lacking the features introduced in later ARM releases it has an unsecured AHB or AXI bus allowing any device that can connect to perform arbitrary DMA (whic
395.
by jsolson 11y ago
Yeah, but, they're actually telling the truth here, right? The iPhone 5C has an unsecured AHB or AXI bus. You can inject arbitrary code via DMA and trigger and interrupt to cause it to run. This went away with TrustZone (present in ARM
396.
by jsolson 11y ago
It's possible; without getting into specifics of any given company's compensation policy, one common compensation scheme involves cash + equity where the equity doesn't begin to vest until some tenure has been reached with th
397.
by jsolson 11y ago
(disclaimer: I work on Google Compute Engine) For queuing AWS has at least SQS and SNS, both of which solve roughly half of what's commonly desired from a queuing system. Google Cloud PubSub coalesces both of these behind a single API
398.
by jsolson 11y ago
(Disclaimer: I work on the hypervisor that lives under Google Compute Engine) 1. PerfKitBenchmarker includes meaningful benchmarks for things like Redis, Aerospike, Memcache, etc. We expect GCE to score well on these when measured in terms
399.
by jsolson 11y ago
Thank you for taking the time to write up the experience. I can think of a few metrics that directly relate to what made it suboptimal. Google is pretty good at improving things when we have metrics we can measure directly. That said, this
400.
by jsolson 11y ago
Virtual CPUs with VTx can be made to run at essentially native speed for most workloads on most fairly modern CPUs. Memory performance can be subpar, although there's public documentation on techniques that can improve that: https:&#x
401.
by jsolson 11y ago
The simplest thing to do of course is just try it. You can install Windows in a VM without a product key and try it without limitation for something like 30 days. That said, "really low latency" is relative. Device assignment will
402.
by jsolson 11y ago
> I wish they had the same exuberance about their support tickets :( Is there a particular ticket (or tickets) you ran into trouble with? I can try to follow up.
403.
by jsolson 11y ago
For me personally[0], the difference isn't being able to speak in terms of marketing, but in terms of engaging with users who have problems. It may not translate to much in terms of run rate, but I like to think (or at least hope) the
404.
by jsolson 11y ago
Update: Actually under rare circumstances we'll validate the checksum on the host side, realize it's wrong, pass it to the guest as CHECKSUM_NEEDED and the guest will happily fail to checksum it and mark it as already validated. S
405.
by jsolson 11y ago
Note that the "crack" in this case doesn't expose the contents of the encrypted flash if the user of the device picked a sufficiently strong passcode.
406.
by jsolson 11y ago
Google Container Engine VMs should be protected from this by the virtual NIC advertised to them. In particular, it advertises support for TCP checksum verification offload which all modern Linux guests negotiate (including the kernel used i
407.
by jsolson 11y ago
Presumably each change in the history will be reviewed, though, right? One assumes the point of doing rebase -i to reorganize commits during review is to present a history with MORE and BETTER context -- one which can be more easily reviewe
408.
by jsolson 11y ago
> Au contraire, rebase can break bisect by creating commits that were never actually tested or compiled. And if your bisection algorithm cannot handle merges, then that's a problem with a bad bisection algorithm. This is only true i
409.
by jsolson 11y ago
> Plus anybody who advocates the use of git rebase --interactive as a sensible response to code review feedback is clearly insane. Why? Prior to a series of changes being pulled into someone else's history you have all the incentive
410.
by jsolson 11y ago
It was, historically, definitely a technical problem. You had to know what the base changelist for a branch was in order to integrate into it from trunk or to merge it back into trunk cleanly. I believe these problems may now have been fixe
411.
by jsolson 11y ago
Maybe. It depends on whether or not the kernel is smart enough to recognize that the pages you've referenced in the AIO operation are pages are mmap'd files rather than ordinary heap memory. It would need to do some amount of page
412.
by jsolson 11y ago
> Seems quite useless to me. For, what if you want to prefix the file with a header first, or add a trailer? Or what if you want to escape certain parts, or zip the file before sending? Or if you want to send the file incrementally? The
413.
by jsolson 11y ago
Assuming you're running multiple pipelines at once you can add robustness in this sort of scenario by running the long-running tasks on guaranteed VMs (complete with live migration on GCE) and the short-running tasks on cheap preemptib
414.
by jsolson 11y ago
As boulos points out ingress is free -- the other question is how much you actually need to egress. Transfer within a GCP region is generally free as well, so unless you really need your giant result set in its raw form, you can potentially
415.
by jsolson 11y ago
First, the actual video content itself does not, as far as I know, stream from AWS. If you watch in the web player you can see the many requests going out to Netflix's CDN. Given that the website is served up over HTTPS, Amazon doesn&#
416.
by jsolson 11y ago
All of those quotes are conditioned on the premise that you're doing research which purports to be relevant to industry. In that context I agree with Matt. > Now, I don't think all academic research has to be relevant to indu
417.
by jsolson 11y ago
That's not really what the article is doing, though... Instead it's saying roughly: much of the work coming out of academia attempts solve problems applicable to industry, but in actuality industry is not actually suffering fro
418.
by jsolson 11y ago
Actually reading the article, explicitly none of those things. Alternatively: if you are doing work that attempts to have industry relevance you should have some idea of what problems are actually relevant to industry. In particular, just
419.
by jsolson 11y ago
The numbers in this are daunting, but I personally believe massively multi-core systems make the problem a lot less daunting than the article makes out. Core counts in big servers can get up over 100 per server for Intel (see Amazon's
420.
by jsolson 11y ago
My current team does not do standups (or Scrum or... any of these things, really). We have a rolling list of open features/cleanups/bugs and a polite triage convention. In my last company we were all-in on what I can only call Scr

More ›