When I got back from the hike I heard on the radio that this location was, ahem, not recommended. It was closed off yesterday and a short while later the crater walls collapsed in that direction.
A bit on the macabre side (but your comments makes me believe you don't mind too much perhaps):
what would happen if you'd fall or otherwise get trapped in lava?
Gollum melted I think but that seems unlikely (it's a movie after all). Left for future archeologists like in Pompeii?
Please don't test on purpose or accident this btw.
it's 1200°C so you burn already before your body even hits the lava. During the Pompeji eruption people got killed by the hot ash (arriving first and travels further) which cooked them alive:
> The individuals in the boat houses died relatively quickly: The volcanic ash blocked the entrance to each structure, and the temperature of the air within probably rose to about 400°C—even hotter than a wood-fired oven. -- https://www.sciencemag.org/news/2020/01/studies-reveal-grues...
I don't think much would be left of a person, but in central Oregon near Bend there's a park called the Lava Cast Forest, which is what it sounds like. About seven thousand years ago or so, there were some lava flows that engulfed live trees. The lava solidified, and what was left of the trees decomposed.
Now, what's left is a bunch of hollow tubes that are castings of the trees. You can still see the texture of the bark.
I assume this sort of thing exists in other lava flows around the world.
A trail around Mt St Helens (Trail of Two Forests, https://www.wta.org/go-hiking/hikes/trail-of-two-forests) has that kind of thing. There is a short tunnel you can crawl through - two trees were laying 90 degrees to each other overlapping when covered with lava. The entrance reminded me of a scene of a typical campy B horror movie (No! Don't go in that cave! The Monster is in there!) I wish I thought to scream to my wife when I crawled in (pretending the Monster got me) -- only thought of it later :-).
There are plenty of modern, distributed RDBMSes that make sharding transparent to the user (E.g. cockroachDB, yugabyte, vitess, many cloud offerings, etc.). Most NoSQL databases end up adding transactions because they are important, and thus the scale advantages for NoSQL systems over relational databases are diminishing, if remaining at all.
Latency at p99 is substantially better at scale with a key value lookup in nosql over postgresql.
Thing is most people don’t have scale these days. You can get a single box with hundreds of logical cores and many hundreds of TiB of locally attached ssd. Until you exceed that you don’t necessarily have scale.
> You can get a single box with hundreds of logical cores and many hundreds of TiB of locally attached ssd
And then that box falls over because the entire region fails like just happened yesterday with OVH. Or it just randomly fails like has happened to me with AWS dozens of times.
Vertically scaling a database on your own cloud instances is amateur hour. Either use a cloud-managed database or one that is highly available.
Well it depends what you consider scale. When you look at the high scale deployments like Slack, YouTube, GitHub, Square, etc they are all using Vitess and MySQL.
YouTube doesn’t use just mysql or vitess these days they migrated to other google databases.
Slack, GitHub both are stupidly shardable. I doubt it’s one RDBMS handling every customer, chat room, git repo. And instead they’ve segmented the workload across multiple instances.
Of course they run on multiple instances. It doesn't work for every use case but it works for more than people realize. Too many people vertically scale, or add read replicas which is a ticking time bomb.
Git is what happens when a technology is born out of hardcore engineering. Most of the world could greatly benefit from its functionality but it’s too hard for non software-engineers to learn. I hope the next era of version control thinks about the problem from a less technically inclined user’s perspective.
Out of curiosity, does anyone know of any good alternatives that make version control easy? I personally love git, but want to see it’s benefits brought to a wider base of people. Projects like Bit are a step in the right direction.
I disagree. "Most of the world" has a hard time understanding hierarchical structure like files and directories. Once you introduce git, which basically adds a time dimension to that hierarchical structure--good luck understanding that. It doesn't matter how user-friendly the tool is if the fundamental concepts cannot be understood.
> "Most of the world" has a hard time understanding hierarchical structure like files and directories.
That is an excellent point that I think deserves expanding.
I submit that files and directories ARE difficult concepts. Pretty much everything is difficult when you look into it enough.
I remember of the time they pointed the Hubble Telescope into a seemingly empty patch in the sky and with long exposure or something, we saw tens of thousands of galaxies from billions of years ago.
Back to the subject at hand, I tried installing gentoo one time and it prompted me for something. I only vaguely remember the word "inode". Here is the first paragraph from the wikipedia from the article on inode:
The inode (index node) is a data structure in a Unix-style file system that describes a file-system object such as a file or a directory. Each inode stores the attributes and disk block locations of the object's data.[1] File-system object attributes may include metadata (times of last change,[2] access, modification), as well as owner and permission data.[3] Directories are lists of names assigned to inodes. A directory contains an entry for itself, its parent, and each of its children.
Files and directories may be an easy concept to understand if you have been exposed to them long enough (not sure how long is long enough) BECAUSE we have a good abstraction. I never had to learn what inodes are or how a filesystem works to use a computer. Can we accomplish something similar with version control?
I wouldn't say so. From watching non-dev people, I can say that they DO know about files/folders and they can very much add the time dimension to them. It just looks a bit different:
Any user who deals with a set of files that need to be coherent, such as a source tree, I'd argue is already most likely technical. Non-technical users are used to having the time dimension, but for individual documents. They understand the notion of revision A and revision B of a document. Most people also most likely understand that if two people edit revision A, they would create separate, conflicting revisions B.
but: most people also accept wihtout fuss, that it's OK that you make a phone call or email and just agree who throws their version out. Or that one user should simply lock the document for editing to prevent this from happening.
"Git for non technical users" (of single documents, not trees) seems pretty easy to build as a layer on top of git. 1) you use LFS and file locking. 2) You "check out" a file for editing which locks it exclusively 3) You completely hide the notion of "local repository" because zero nontechnical users, ever want to be bothered with distributed version control. Basically - you make a dumber but more robust verison of SVN on top of git-lfs.
Even just using text files surely? Even a single document can benefit from git-- or are you thinking that tree-history and other things which are sometimes provided by editors are enough?
If you can live with non-Git then Pijul[0] seems quite promising both in terms of UX, but also power relative to git. (It's probably still quite rough around the edges right now, though, it's 1.0.0-alpha).
There's also Fossil[1] which is stable and seems to have pretty good UX.
Unfortunately git is the easy one, but you're right about the accessibility. I headed a transition from svn to git once, and even though I think git is conceptually simpler overall, it quickly became apparent who paid attention to the graph theory portions of Discrete Math and who didn't.
Pijul [0] also claims to be "an intuitive VCS unlike git". The idea of composing patches is interesting, but I haven't got around to actually trying it yet.
I’ve always been curious, if you’re working on a really complex system with lots of disparate services (or even if you are using a managed database like Spanner for example), what does your development environment look like? Do you spin up containers for all the services? Run a compatible RDBMS instead of the managed database? All my experience has been with systems that can be set up locally - how do you go about developing/testing/debugging without that?
This is one of those posts where you can really feel the value of senior engineering/previous experience.
I definitely have not approached choosing a new technology with the velocity vs. maintenance trade off, instead just choosing the technology best fit for the job at hand. But when looking at a system holistically, this may not be the best choice. It’ll be good to at least know to consider this in the future (although I’ll admittedly probably still bias towards “fun” technologies).
It was done right before the first Crew Dragon launch. They do such a great job of digging into a company’s history and the business model that powers them.
100% agree with SpaceX, mine 2nd would be Pinduoduo and how revenue and cash held can be completely different and how advanced China is many areas (like logistics) compared to the west.
https://www.acquired.fm/episodes/pinduoduo
People often argue that remote working flexibility is the future, but I think it will be detrimental to both younger employees (losing out on mentorship) and those that do go into an office.
An office is at its most effective when everyone is there to collaborate; it loses most of its benefits when only some people are in person. Because of this I think there will be many companies advertising themselves as office-only as a benefit.
I think it can be, but only of your business is monolithic. My work affects and spans offices in 3 time zones, and very disparate ones at that. I have a few local colleagues, but I can’t think of a situation where talking over the wall or desk has been more effective than a slack message or a call.
Fair enough. My main work experience is at a smaller startup that has a heavy focus on collaboration so my opinion is biased towards that. In my experience trivial things can be solved with a slack message/call but I greatly prefer in-person interaction for more in-depth debugging, etc. But I understand how that’s not necessarily applicable to all situations.
My main concern, and the reason I posted the above comment, is I think companies may over -correct with hybrid work models. For those that prefer office work a hybrid work model is barely better than a remote-only culture; the office loses its benefits when everyone isn’t there. I’m not arguing these remote-only/remote-first cultures shouldn’t exist - in many cases they provide large advantages - but think there will be a significant desire for office-only environments for those that prefer office work.
I’m not arguing you should. But there are undoubtedly some people who would. And a hybrid work environment is detrimental to those who want that in-person experience. There should be opportunities for remote-only/remote-first cultures and in-person only jobs going forward. My argument is that a hybrid model misses the point - you lose almost all the benefits of in-person work in the process.
EDIT: I’m a new employee as well, just so it doesn’t seem like I’m projecting what new employees would want without at least some personal experience
For the past few months I’ve been taking markdown notes which are then transformed into HTML and pushed to an s3 bucket for public viewing. Probably not the best solution within this thread but it costs approximately nothing is a great way to get access to a web-based version of your notes anywhere.
This allows you to throw extra features in the HTML like searching for files and randomly selecting a note for viewing at your leisure.
I love the idea that I have a place to publish random stuff that I don't have to spend lots of time writing, I can just drop a sentence and maybe someone else will find it useful.
The big problem is that Google will basically never give you that page in a result, no matter how relevant it may be to your query, because it prefers SEO-rich content farms like Wikihow. I was thinking that a service would at least be easier to search/remember to go to.
Zoom actually is profitable and was one of the rare technology companies to be profitable at IPO. Although they do offer a lot of free services they nailed the freemium line - 40 minute free calls - since the most effective business meetings are 45 minutes.
