Xorlev

11 ms·

91.
by Xorlev 6y ago
Speaking as a corporate programmer working from home indefinitely, I'd kill for IDEA Viewer. My laptop can't have code on it, so I'm forced to remote desktop to my workstation to use IntelliJ and CLion remotely, which is a mi
92.
by Xorlev 6y ago
There's probably an npm package for that. :)
93.
by Xorlev 6y ago
I was surprised to see their canary stages are just 5 minutes. Many problems take longer to manifest. That seems like a fairly risky release process.
94.
by Xorlev 6y ago
That also means when it does go wrong, it takes much longer to fix. Good operational practice is to decrease MTTR, not make it worse.
95.
by Xorlev 6y ago
It is! https://github.com/Creality3DPrinting/Ender-3 It uses http://marlinfw.org/ with only modifications to the configuration (which, you need to do for any printer using Marlin). But the above link a
96.
by Xorlev 6y ago
Totally agree that enthusiasts are overrepresented, but that seems like a leading indicator, at least in the performance and gaming market (which is a growing portion of the overall desktop market).
97.
by Xorlev 6y ago
https://support.google.com/a/answer/9381067?hl=en Approvals ends up solving some of this by freezing changes to approved documents.
98.
by Xorlev 6y ago
Correct, only editors can see edit history. Making a copy also takes a snapshot of the document without history.
99.
by Xorlev 6y ago
Again, depends on if you're doing a lot of IO, especially if you want request-based IO parallelism. It's worth mentioning that even if you're IO-bound at your DB, running an async application server now means you don't n
100.
by Xorlev 6y ago
Slowing down actuation of prod changes to be over hours vs. seconds is a far cry from the large org / small org problem. Ultimately, when the world depends on you, limiting the blast radius to X% of the world vs. 100% of it is a signif
101.
by Xorlev 6y ago
$0.90 in 1999 is $1.41 today. Near me, gas is ~$2 (give or take). So, more expensive (~40-50%), but not double.
102.
by Xorlev 6y ago
Note: this starts at noon PST.
103.
by Xorlev 6y ago
Did you mean Hadoop/HDFS? HBase works pretty analogously to Bigtable (though, having run HBase in production at $dayjob-1, I'd take the latter any day).
104.
by Xorlev 6y ago
I know this was intended to be a joke, but it didn't really land with me.
105.
by Xorlev 6y ago
It's not a secret that we use Borg, you can read about it online.
106.
by Xorlev 6y ago
Exporting via Takeout will retain the geolocation tags. The API isn't the best place to export.
107.
by Xorlev 6y ago
Paint et. all work too, what an incredible simulation! :)
108.
by Xorlev 6y ago
There's a lot of reasons TPUs could be poorly matched for your workload, including model complexity (or lack thereof), the way your model is setup, the inputs to your models (if you're bottlenecked on IO, including host <->
109.
by Xorlev 6y ago
Totally agree! And for what it's worth, sorry it didn't work out. My reply above was to the parent who seemed to think it wasn't all that useful. I looked through the code and found a lot of goodies in there. :) If nothing el
110.
by Xorlev 6y ago
Does code released have to have commercial applicability to other organizations to be relevant? It seems like a post-mortem code dump.
111.
by Xorlev 6y ago
RTT between VMs in the same GCP zone is often well below 100us. Are you also measuring the latency of the query itself? I realize not all networks are built the same, but it seems like your benchmark case is much worse than what's poss
112.
by Xorlev 6y ago
Disclaimer: I work for Google, but not on Chrome or Search. There's plenty of worry to go around. I would like to put aside the (very understandable) fear about Google for a second because I think the GP makes a good point. My parents
113.
by Xorlev 6y ago
Your statements are overly hostile and hyperbolic. There's been plenty of great software out of Google and lots of good ideas taken for granted. I'll give you folks trying to build things the Google way without Google tools or sca
114.
by Xorlev 6y ago
https://github.com/cloudflare/quiche
115.
by Xorlev 6y ago
> Additionally, these trackers are usually picked out by business analysts / product managers on different teams This. I remember at my last company, we ended up with 8 or 9 different analytics tools all getting different data and s
116.
by Xorlev 6y ago
> For myself I have used one of the popular quickly and cheaply grown accounts to slide into the DMs of local women. :|
117.
by Xorlev 6y ago
You don't know that. For all you know, some kind of knowledge graph tag propagated too far. There is zero evidence of it being intentional. You really think they'd just erase the photos of UK PMs? Please. If there was some kind of
118.
by Xorlev 6y ago
Or, perhaps it's a bug? You're ascribing a whole lot of intent to something without any details to support that position.
119.
by Xorlev 6y ago
Rollouts often happen on a per-country basis. Including bugs.
120.
by Xorlev 6y ago
Maybe don't be so alarmist? Bugs do happen.

More ›