clord

For branching schemas, yes it gets complicated. But at least you get a nice conflict while merging the branches (at the right place) These conflicts help guide a rewrite of the schema history into something consistent. That's part of it: when a merge happens, you basically have to create a new migration that does the merge too.

and if you never merge the forks, there is no problem. they likely won't share a production database anyway, right?

I wrote[0] about a method I've used with success in the past. Essentially you use the previous version's hash as the name for the next version. The benefit is that merging the work of multiple developers is easier. I ran the process manually when I did it and would love to hear if someone writes a script that makes the process easier.

[0]: https://medium.com/@clord/for-migration-of-schemas-use-versi...

> A natural image is formed when a collection of objects is illuminated by incoming light, and the resulting image is projected onto the retina.

But that's sort of the point no? The incoming information is not an unstructured white noise of photons striking our retina. There is a sort of structure to the information that can be modelled. One such model of this structured information is as a "grammar tree" (really just a tree, we're coders here.) The example in the article is that the arm occludes the teepee, which occludes the background trees. Any visual system needs to break this hierarchy down.

This article cuts to the heart something I've wondered for a long time.

The common advice in all the classic texts is that developers should not roll their own crypto because smarter people have thought of more vulnerabilities and addressed them in battle-tested code.

But news like this shows that there is an antithesis: the conventional encryption techniques are also potentially widely exploitable by state-level actors. Furthermore if I was someone holding solutions to cherry-picked primes for well-understood algorithms in wide use, I'd be complaining loudly every time someone wrote a bespoke library too. I'd be paying to publish books that recommend no one write their own crypto because it's just such a darned hard problem, especially with so many high quality alternatives out there tested and ready to go.

Granted, one should certainly have a repulsion to to writing custom crypto for all of the many good reasons, but it makes me think it's worth putting in more than the minimal effort into it, especially when lives are on the line.

The article points out this can cover huge swaths and millions of buildings at once. It speaks of covering a continent with subsidence alerts. Afixing gps to each building might not be competitive.

You dislike the particular compromises made to implement a bad version of the ideal. That does not mean the ideal is to be avoided, just that it is more complicated to achieve than the OSX team was willing to implement.

Yes, I do this too, and it angers me because it proves that this datapath is possible to build. Double-dongle sucks, but it works. I marvel at how backwards apple can be when it wants to be. I wonder if this new headphone dongle they announced can be hacked to do charging.

Are you suggesting there is no room for quantum effects in the brain? Honest question because I thought determinism is out of fashion now-a-days for that reason.

... but done with proper consideration for privacy.

There should be (is there?) an OS designed around zero copy principles. I want to get packets from the network delivered into a block of memory, and then all subsequent parsing and reading occurs on that memory, with OS provided framing and fine grained policy determining how things are interpreted by userspace.

Something like the memory trick used in this virtual ring buffer but far more powerful.

There will always be experts (and journalists) who can't think of the impact of some change. But I believe tropic cascades are a real thing in food webs, and we can't really know for sure that something will have no impact until we do the thing. If there is one thing we're learning right now, it's that these cascades are real and quite dangerous. Now that this is known, no one should in good conscience set off a possible cascade without knowing something about what might happen.

Humans can improve on nature once they understand the impact of an action well, so I'd get behind a well funded study into what effect getting rid of mosquitoes would have — one with an experimental basis and not just a journalist's thought experiments.

Suppose the market just tracks widget factories. Every factory produces the same quality widgets, all shipping is flat-rate by distance and there are no tariffs or taxes. In this market, it's easy to determine market value of any particular widget factory: market share of world-wide production.

Suppose everyone expects the demand for widgets to grow by 1% per year for the next 10 years. This growth might get priced into the market with some sort of net present value calculation. Capital will flow into building new widget factories to fund all of this new construction.

Now what if actually an asteroid hits a city with lots of widget demand. 50% of demand is wiped out instantly. Well in this case, everyone was overvaluing the market for widgets! All of those net-present-value calculations were way off. The market will correct.

Or maybe the asteroid takes out 50% of production. Suddenly much more investment in widget factories is required. The price shoots up to capitalize all of the construction, since demand has not changed.

In either case, before the asteroid the market was not correctly valued since no one had yet priced in the asteroid.

In one sense yes — if the sum total of the market is '1' and everyone is just some fraction of that. Then there can only ever be relative price changes.

But the sum of all prices does in fact vary. I'm sure you're familiar with Tulip Mania. Prices can be bid ever-higher in a cycle, and they can also decline in tandem as in the depression.

For the entire market to have an expanding total price, new money has to be entering the market, or the net real value of the underlying financial assets has to be declining. Either way, the entire market can become overvalued or undervalued, especially compared to non-financial (or illiquid) assets denominated in the same currency — e.g., wages, energy, or land.

I'd rather have each module have its own dependencies as subdirectories, sort of like node_modules. Everyone resolves paths from cwd.

If someone wants to share the same jquery across subdirs, I'm sure http redirects could be configured. Or alternately some sort of corporate versioning convention that developers can expect to have with absolute paths. Certainly better than ../jquery.

a cache drive is l2arc. Brendan Gregg's post[0] explains it well, but the TLDR is that l2arc is very much like an l2 cache in a cpu. Recent and relevant items are held on a fast drive in case it will be needed in the near future. It is really an L2 cache: L1 is called ARC and is held in dram.

If I recall correctly, the extra indexes and tags mean l2arc requires more memory to work efficiently.

[0]: https://blogs.oracle.com/brendan/entry/test

Instead of rebooting, it should (in theory) be possible to use the kexec syscall to replace the kernel. I would rather just reboot for various hardware-related reasons though...

Could it be that Alzheimer's plaque is the ultimate expression of an attempt by the body to deal with CJD?

From what I understand, it has cleared the space but there are still harmonic crossings that can't be cleared. Interesting challenge to the new definition - must the orbit be clear, or only clear sailing through the orbit.

Computers knowing about focus is quite a compelling future direction. There are lots of really good possibilities. But also a few problems.

For instance, If you can track focus, you can make the monitor seem larger than it really is. Just scale everything that isn't being looked at down a bit. As the gaze shifts towards other objects, shift things slowly around and overlap the enlarged window over other background windows.

You can make focus-dependent shortcuts. Imagine vim with nouns and motions that can refer to and act on the focus point. `ytF`: yank-to-focus, where F is a motion from cursor to focus. Lots of rich possibilities there.

The only major problem I see is that you can't really share work easily (unless you open it up to multiple focus points somehow.) Also, it would be frustrating to have to look where you're typing. I'll often look at something else while typing just before switching tasks. I'm going to start paying attention to my focus more to see if there are other potential pitfalls with the computer knowing about it and changing modes in response.

In general I like the possibilities opened up by having focus. Heck even with a traditional mouse+keyboard, the extra data would help the computer understand us better. It might be more suited to a VR desktop, where the 'multiple-viewers' problem does not exist.

> iPhone production can increase as much as demand asks for. If we let anybody build one, there'll easily be an iPhone for everybody that wants one.

What? This statement makes no sense. Here, let me demonstrate by quoting an ancient islander who likes to build statues:

> Tree production can increase as much as demand asks for. If we let anybody grow one, there'll easily be a Tree for everybody that wants to cut one down.

It's absurd! and exactly the thinking that will lead to people cutting down the last tree. iPhones depend on resources just as much as those trees, if not more.

Actually it's a good name. C got incremented, and we keep original C locally while some weird side effects no one really wanted keeps happening in committee until the end of time.

On a local disk, which is mirrored periodically to a redundant ZFS filesystem in a utility room.

Just started abandoning Photos today too. I have a 1.5TB library, so it never did fit on iCloud Photos. Switched back to plain old file system directories with all the files named uniquely, and it's surprisingly refreshing. No beach-balls, no BS prompts about not having enough space on iCloud even though I don't use it. Better integrated with the tools I use too (DxO).

Apple's whole 'Each app is it's own file manager' theme is getting really tired.

hopefully autoprefixer-cli is fixed. My makefile builds have been failing randomly and unreproducably since they split it out of core.

Been using ledger for 9 years as well (hledger lately.) Never have a hitch. The only complaint I have is that I can't (easily) keep AR in sync with online invoicing software. minor though. Would love to have some git-style integration with my bank and other accounts. `ledger pull savings` or `ledger pull invoices`

No mention of beneficial microbes, which is what I imagine the primary benefit of this practice is. Soil is loaded with the local microfona.

The biggest problem I see with this is that we will lose the incentive to dial down extraction of fossil energy. The non-warming side-effects of oil and other deposits such as carbonic acid in the oceans would become just as severe if not moreso.

I think the Mill team is a good acquisition target for Apple... Apple could swiftly switch their whole stack over and gain remarkable benefits. I can't think of another company that would even be able to switch to the mill line.

Cane toads are introduced. Are you telling me Australia has no ants?

The reason from my perspective is that it is different than a traditional unix system, and what unix emulation it offers is limited (to put it in perspective, it feels like a bad vim emulation... your native skillset doesn't quite translate). Lots of issues with dealing with data formats (EBCDIC). Access to open source compilers and scripting languages was limited. Granted these were strengths of the platform as well, but it made our workload (maintaining a compiler for the target) harder than other unix-alikes for those of us not fully versed in JCL.