HNHacker News
TopNewBestAskShowJobs

fusiongyro

5,495 karma · joined September 17, 2008

Daniel K Lyons

http://www.storytotell.org

Feel free to email me at fusion at storytotell dot org.

[ my public key: https://keybase.io/fusiongyro; my proof: https://keybase.io/fusiongyro/sigs/aYrkbCZouBBhqp_qOKPGdBfIn---A8P8wPVCKLSlpnA ]

submissionscomments
fusiongyro··on Almost everything on computers is perceptually slower than it was in 1983
I agree with you about false machismo. I'm trying not to say that hard-to-use tools are essentially good. I am saying that some essentially good tools are not very new-user-friendly, and that worrying excessively about the new-user experience has been detrimental overall.

One reason for that is the "new user experience" for a more powerful system will imply learning some things that are not directly relevant to whatever problem you are trying to solve. Excel worksheets, layers in Illustrator, "command mode" in vi and Blender, etc., are more like gas/brake pedals than they are like "do you want directions to this destination?" I don't see how you can avoid thrusting some learning on new users of a powerful system. Powerful systems are going to have powerful metaphors that are probably not totally intuitive at first.

You can do more stuff with Illustrator than Omnigraffle. It's a more powerful tool. That brings with it a longer or steeper learning curve. But, if you just want to make a few charts, Omnigraffle is easier to use.

I do feel that, over time, we have become very gun-shy about making complex applications that have richer models like this though, preferring to make applications with simpler UX that aim directly at smaller problems. And I think this is partly because UX is more straightforward for smaller problems.

fusiongyro··on Almost everything on computers is perceptually slower than it was in 1983
There is a big dose of fashion and fad-following in usability. The things you mention are manifestations of poor decisions made in the name of usability, but actually in the interest of being pretty.

I think even making the right decisions in terms of usability, tethered to a few short-sighted user stories, will result in this situation. Usability should be in service of a higher purpose, which should be enabling the user to understand how to use something powerful and flexible that does not constrain them only to what is envisioned by the developer or the UI architect. That's what has happened here with Maps. You can perform User Story #1 or #2 with zero friction. But there is no way to do what @gravislizard wants to do, which wasn't anticipated by Google or specifically designed for.

fusiongyro··on Almost everything on computers is perceptually slower than it was in 1983
Yes, exceptions are exceptional.
fusiongyro··on Almost everything on computers is perceptually slower than it was in 1983
In the name of usability, we have practically neutered these computing machines. I think optimizing for user experience is partly to blame for this. You don't want users to have a confusing experience the first time they use the application, and you want the application to look good and be inviting. But that may be at cross purposes with making applications that allow open-ended exploration. Exploration implies that the user is driving the process, which suggests that the UI is not actively trying to constrain them to the usage patterns it is designed for.

I think pretty much everything in computing is like this. By setting our sights on enabling users rather than training operators, we have not only limited what they are able to do, we have actually turned ourselves into users, and we're basically unable to program without the help of frameworks, libraries and Stack Overflow.

fusiongyro··on If you type the letter “i” and it autocorrects to an “A” with a symbol
I agree, and I don't need millisecond precision on this, but I think multiple seconds is absurd.
fusiongyro··on If you type the letter “i” and it autocorrects to an “A” with a symbol
My favorite issues I've noticed so far:

- Messages scrolls stupidly so I am looking at two exchanges ago and can't see the most recent reply

- Timer alert is delayed ~10-20 seconds if you look at the screen as it hits 0

fusiongyro··on What If OpenDocument Used SQLite?
Is there a way to reformat XML without introducing space?
fusiongyro··on Netflix Employees Are Happier with Their Job Than Facebook or Google Employees
I guess I'm giving you a tip o' the hat as we pass each other by on our respective hobby horses.
fusiongyro··on What If OpenDocument Used SQLite?
Was it Fossil SCM by chance? Because it happens to have the same author. :) http://fossil-scm.org/
fusiongyro··on What If OpenDocument Used SQLite?
The tool you're describing is `xmllint --format`.
fusiongyro··on Netflix Employees Are Happier with Their Job Than Facebook or Google Employees
The rest of my point is basically "on their junior developers." What I'm referring to is the mismatch between what Google and Facebook actually do (monetize machine learning to sell ads) compared to what they claim to do when they recruit ("come change the world" etc.). The same is not true of Netflix with DRM.
fusiongyro··on Netflix Employees Are Happier with Their Job Than Facebook or Google Employees
Without even looking at the article, I would guess so simply because what Netflix purports to do is what they actually do and how they make money, so there's no ethical shell game being played against starry-eyed junior developers.
fusiongyro··on What are some of the most disliked programming languages?
I want to agree, but C is even older than C++ and is definitely not in the same position. Ultimately, each language is different and the way they grow is different. C++ has just never been all that conservative about admitting new features. Their primary concern seems to have been the effect on the runtime performance of the binary, not whether the feature makes sense or mixes well with the existing features, or the effect on compilation time.
fusiongyro··on What are some of the most disliked programming languages?
This is probably why Stroustrup famously remarked "there are two kinds of language: the kind people bitch about, and the kind nobody uses."
fusiongyro··on Catalan parliament declares independence from Spain
Is there a legal way for US states to secede?
fusiongyro··on Catalan parliament declares independence from Spain
Point #1 trails off, can you finish it?
fusiongyro··on I have no side code projects to show you
I agree that it shouldn't cost you a job if you don't have side projects. It shouldn't guarantee you a job if you do—like any other demonstration of who you are, it should cut both ways. I always ask to see them if someone has them, and it's hurt in about as many cases as it has helped. If you can't program but have some side projects up on github anyway, I will be able to tell.
fusiongyro··on Coda, a “next-generation spreadsheet” – from rows and columns to custom apps
I like it quite a bit myself. But I don't make heavy use of spreadsheets.
fusiongyro··on Coda, a “next-generation spreadsheet” – from rows and columns to custom apps
Apple's Numbers spreadsheet has this feature today as well.
fusiongyro··on Games Look Bad: HDR and Tone Mapping
I think this takeaway misses a significant point in the article: even a picture of that display will look better than it does on the display itself:

> [W]hy does the LDR photo of your TV look so much better than the LDR tone map image? There’s another tone map in this chain which nobody thought to examine: Nikon’s.

The author isn't merely saying that games are too colorful. He's saying the colors are wrong from both an artistic perspective and from a utilitarian perspective. He's saying this is because the people making decisions about how color should be rendered are either ignorant of how to make the decision (copy-paste code) or make the decision without real awareness of how it will effect either of those properties.

fusiongyro··on EtherDFS – The Ethernet DOS File System
Not a problem most of us have in 2017. I wonder what the use case is for having a bunch of DOS machines.
fusiongyro··on ESO Telescopes Observe First Light from Gravitational Wave Source
http://jobs.jobvite.com/nrao/jobs
fusiongyro··on ESO Telescopes Observe First Light from Gravitational Wave Source
I work for the NRAO and concur with WD-42: there are more positions than software developers. The main problem is finding people who can withstand our sadistic hiring process and then not laugh when we tell them the salary and where you will live. :)
fusiongyro··on Killing the C++ Modules Technical Specification
Adding to your point, there are two string types you should use in Haskell (ByteString and Text) and neither of them are the type you get from an unadorned string literal (absent OverloadedStrings and some expectations).
fusiongyro··on PostgreSQL 10 Released
This is why almost all ORMs create a transaction and have a "unit-of-work" that encompasses a single transaction.

Race conditions are always surprising. Fortunately, we have a simple remedy for that: transactions. :)

Suppose you have another connection and it runs DELETE FROM t WHERE id = (SELECT MAX(id) FROM t). If that winds up in the middle, it could screw things up for you. Is it likely to happen? No, but again... the remedy is simple. Also, the scenario you describe seems simple enough, but what happens when you bulk insert a few thousand rows?

By the way, with Postgres, there really is no way to not have a transaction. If you don't issue BEGIN, it treats the statement as a small transaction. So you're getting the same overhead on a single statement, if your plan was to somehow avoid the overhead.

If this makes you question the database's usability, I have some bad news for you: all ACID compliant databases work this way. The "atomic" unit is the transaction, not the connection. The consistency guarantees pertain to the transaction, not the connection. This is not a weird thing about Postgres, this is how they all work (except, of course, that MySQL can't always roll back aspects of a transaction, like DDL).

fusiongyro··on PostgreSQL 10 Released
There's a difference between not using transactions because you don't know better and not using transactions for reasons that arise out of a deep understanding of what is going on. 99% of the time, it's the former. The argument that transactions aren't perfect is a shitty reason not to use them, the kind of bad advice that doesn't hurt you today, but may hurt you in the future.
fusiongyro··on PostgreSQL 10 Released
You always need a transaction, even just for reads. Otherwise you are not reading from a single consistent snapshot of the world.
fusiongyro··on PostgreSQL 10 Released
I balk at the added complexity. Just use two queries. This is why God created transactions.
fusiongyro··on PostgreSQL 10 Released
I'm having trouble imagining why you need this. Often when people come up with "interesting" scenarios involving UPSERT, they are trying to use the database as persistent memory for some algorithm, rather than as a relational database. What are you actually trying to do? Why would it be insufficient to do something like:

    BEGIN;

    INSERT INTO entities (name)
    (VALUES ('a'), ('b'), ('c') EXCEPT SELECT name FROM entities);

    SELECT id, name FROM entities WHERE name IN ('a', 'b', 'c');

    COMMIT;
fusiongyro··on PostgreSQL 10 Released
Migrating to Postgres will certainly be significantly more work than going to MariaDB for you. But I would still recommend you consider it, because Postgres is safer, more powerful and has fewer edge-case behaviors. If your systems are highly-wedded to MySQL's idiosyncrasies, the difference in cost may be too high for you to do it right now, but I would seriously consider it.
← PreviousPage 7 of 34Next →