I wouldn’t be surprised if they counted those as well. I don’t think they’ll go so far as installing cctv inside the bathrooms lol
328 karma · joined January 25, 2017
I can make pretty much the same argument for say, vim or emacs, and say that plain markdown + fugitive/magit hits the sweet spot.
I am just trying to understand where's the vendor lock-in thing coming into the picture here.
Wait until you start reading news ... oh wait, elfeed ... never mind.
Well, you get the point ;-)
Just don't install a kernel - emacs already comes with an OS :P
So to answer your question, yeah, I fully expect my employer to record what I am doing all the time and it’s a price that unfortunately I have to pay to survive in this digital world.
The person in question really did start a new party that was not coupled with any of the existing parties and won the capital state with crushing majority.
That is a sign of a working democracy in my book. I am not sure if US is one of the better version of a well functioning democracy given its two party system. I recall that it was one of the things that the founding fathers were afraid of when they were debating the constitution.
Is it because you’re assuming a cloud backup is necessary component of syncing solution? In that case, you can just designate one of your synced devices to keep backups to S3/B2 :D
<side>must resist the username checks out comment</aside>
This trivia about Nitrogen used to be my favorite when I was studying chemistry in the school. Like there’s lot of nitrogen in the atmosphere and oxygen but they don’t “burn” because (almost all?) nitrogen oxidation is endothermic!
Mindfulness should focus first on observing and not judging/labeling things as good/bad. That just leads to more misery!!
The instructors should certainly realize that mindfulness and it’s experiences are very subjective and each individual has different experiences. One must be careful not to compare two students and say one is a better example of mindfulness than the other.
Use the great suspender extension; my friend in academia thanks me every single time we meet. And now I see them have 4 chrome windows open with each having ~20+ tabs in it!!!
For them everything happens in browser, its synced and the great suspender lets them run that monstrosity without choking their MacBook Airs
I an not saying it’s right or wrong. I think it’s the most practical and thus most pervasive use of docker.
The company and the project provided a great service by creating lot of shipping/container related metaphors though. That’s very much invaluable in my book. Providing thousands of developers means to talk about objects that are static and dynamic from an architectural perspective.
And you have tools like Jamf for any enterprise fleet management needs
If I may ask one more question, would you happen to know if the authors or other researchers who are entertaining any theoretical work on the experimental design and training methodologies of GPT/BERT? As in why does it work? What is the significance of training via the "fill-in-the-blanks" method?
Don't get me wrong - the work is great and the SOTAs are amazing, I would be just happy to have a chat to discuss and bounce some ideas what all this means and why do these methods seem to be working so well. Papers/articles/blog-posts are always a pleasure to read!
Clearly, the authors have given careful considerations to the issue of contamination and have provided reasonable analysis and a careful argument regarding over fitting the existing benchmarks.
On the other I was wondering if the authors would like to consider purposefully creating a type of "out of sample data" for "creative evaluation"? Of course, GPT is no stranger to creativity, so it would be a fascinating challenge to come up with methods to create such datasets that are truly creative and challenge GPT-{N} to prove its mettle.
For example, would it be possible to engage a really good creative writer* along with a highly experienced school teacher to take on the Reading Comprehension task and create few "tricky" evaluation samples that not only go above and beyond the contamination objections but also challenge the human intelligence to be careful not to fall into common traps?
This way lies a different evaluation metric - a subjective one perhaps, but it's a start. Just a thought experiment - that's all.
* so that they can come up with new ways to trick GPT/humans a teacher knows the common mistakes the average student makes
Edit: Duh, my head immediately screamed GANs the moment I pressed submit, lol. But I am not sure if GANs make sense for NLP tasks. Like do they make sense if humans/domain experts try to solve them?
If there are Billions of parameters in the SOTA models, how do we argue that they are not over fitting?
Similarly in this case, we’re reasonably confident that London is a place on earth with a property of time that is being asked. But without the additional context (or supplementary logic) we can’t know for sure which London it is: Canadian, American, or English.
Until we provide such context, London is in a superposition of all the possible meaningful state.
I mean you can try to uninstall it. But apparently it comes back.
Also, thanks for the ERPNext.org suggestion - sometimes I wonder how many open source gems I have yet to see. (No, please don't link me to one more awesome-X list :P)
Most people here seem to be making a self fulfilling prophecy of keybase's death.
But I like to think that Zoom intends to reuse large parts of keybase codebase:
> Logged-in users will generate public cryptographic identities that are stored in a repository on Zoom’s network and can be used to establish trust relationships between meeting attendees. An ephemeral per-meeting symmetric key will be generated by the meeting host. This key will be distributed between clients, enveloped with the asymmetric keypairs and rotated when there are significant changes to the list of attendees. The cryptographic secrets will be under the control of the host, and the host’s client software will decide what devices are allowed to receive meeting keys, and thereby join the meeting. We are also investigating mechanisms that would allow enterprise users to provide additional levels of authentication.
Will the founders be interested in releasing parts if not all of the server code to the public? I believe the founders' mission is still achievable and can be carried out, should they be willing to release the code in public.
Then, you can slowly try to switch one thing at a time, for example start editing all Dockerfiles in spacemacs, or maybe do that one hello-world project/app in clojure/phoenix using spacemacs. See how you like it, tweak it a bit.
Finally, decide on using it as a daily driver for a new project and stick with it. It will grow on you.