HNHacker News
TopNewBestAskShowJobs

erezsh

327 karma · joined May 18, 2009

Skilled Programmer

My blog: http://blog.erezsh.com/

My github: https://github.com/erezsh

submissionscomments
erezsh··on Ask HN: What weird or hard problems are you trying to solve?
I'm trying to replace SQL by building a language that compiles to SQL. It has first-class functions, nicer syntax, better type-system, introspection, and other things you would expect from a modern language. But in the end, you still get SQL's performance, and the ability to use it with dozens of database engines.
erezsh··on Scaling Pandas: Comparing Dask, Ray, Modin, Vaex, and Rapids
Do you also have experience working with SQL databases? If so, how do they compare to Pandas in terms of performance? (with or without these extensions)
erezsh··on Foam – A Roam Research alternative with VSCode, Markdown and GitHub
I think the problem is that you're including these links in the README. New visitors expect all the links there to work, but they are broken.

If the README only included basic information and a link to the github pages, it would be easier to follow.

P.S. since I'm already typing, have you considered supporting semantic links, in addition to the regular ones? It will make the relationships between the pages clearer at a glance, and allow for better automation in the future. It might also help in visualization, for example you could toggle only specific relationships, to get a simpler graph of the information you care about.

erezsh··on Foam – A Roam Research alternative with VSCode, Markdown and GitHub
What you're describing is essentially a recommendation system with an understanding of natural language. There's a lot of research going into it, but it's still a very difficult problem to solve.
erezsh··on America is giving up on the pandemic?
> Average daily deaths in US caused by corona virus: 1000

Just wait a couple of weeks...

P.S.

It might already be higher: https://www.ft.com/content/6bd88b7d-3386-4543-b2e9-0d5c6fac8...

erezsh··on Ask HN: Dear open source devs how do you sustain yourself
Looks nice. Have you considered creating a free-tier for open-source projects?
erezsh··on Ask HN: What's your quarantine side project?
It's a cool idea, but you should ask a lot less permissions. No way am I giving you permission to make changes to my account, like follow/unfollow people for me.
erezsh··on Show HN: A modern BASIC-like teaching and learning programming language
'write' is probably a better term for that
erezsh··on Ask HN: Great fiction books that have had a positive impact on your life?
I agree. I read it in English and even though I felt some of the punchlines, for the most part I had no idea why they were supposed to be funny. I assume a lot of it is cultural references.

Catch-22 is great though. Dystopic and hilarious at the same time.

erezsh··on Show HN: UniversalDataTool – open-source collaborative data labeling
Does it have an API for adding labels programatically?

I couldn't find any documentation, or even installation instructions..

erezsh··on Using Anki to remember what you read
I don't see how cards would help me with mathematical ideas. You have to get them on a fundamental level, and then they're hard to forget. I'm more likely to forget the name of the theorem. But space repetition might be useful for formulas.

But I'm not too crazy about spaced repetition. I used it to learn all the capitals in the world, and after about a month I managed to do all of them without error. But after two years of barely using the knowledge, I only remember half.

Of course, I could keep reviewing it every few months to keep it fresh, but that's just not an efficient system, when you think of how many things I'm supposed to (and do) remember.

For me, at least, the best way to remember things is by tying them in as many associations and metaphors as I can, and that gives me a pretty reliable recall. It has some downsides: It's a bit more work than spaced repetition (requires creativity, for example), and it's not that good for unconnected data (but then, are random facts that useful anyway?). But I think that as a long-term method, it's much more solid.

erezsh··on All Programming Languages are Wrong (2018)
It's not contradictory. I'm saying that the things he says are true, and there are many languages that follow that line of thought.

Yes, faster than C++. Better abstractions can, sometimes, give the compiler more flexibility to optimize.

erezsh··on Snel: SQL Native Execution for LLVM
Hmm, interesting concept, and an article is nice, but.. why not release the code?
erezsh··on All Programming Languages are Wrong (2018)
The author has a good point, although he miscommunicates it: The common use-cases, such as servers, business logic, glue code, etc., don't usually require hyper-optimized performance. It only needs to be reasonable and predictable.

They need a language that is easy to work with, and share with others, and there are few that really do that.

But of course, there are. Go, Python, Julia and others I'm sure, all prioritize comfort over performance. Although, as compilers improve, they might one day actually become faster -- since the user writes more abstractly, there is a wider search space for optimizations.

erezsh··on SQLite Is Serverless
My project is an interpreted query language that compiles to SQL (with support for several backends).

It is more capable than ORMs, and provides a layer of abstraction that SQL direly needs but lacks, as well as a shorter syntax, that is in-line with other popular languages.

Here is a very early version of it: https://github.com/erezsh/preql

I've kept working on it, but privately, and I'm trying to make it into a product.

I will probably release it as open-source when it's ready. I still need to figure out the right license, financial model, etc.

erezsh··on SQLite Is Serverless
Well, many choose to do their data-science over aws and similar, so I'm not sure there's a big different. I see you point about throughput and network load, but part of DS is data analysis, where the work is mostly exploratory: finding connections in existing data, and working heavily with aggregated data and previews, rather than just using it as a pipeline for other systems.

I think for most JAMs the network is a bigger hindrance than the query time. So, I hope you know what you're doing.

Anyway, cool project. I'll make sure to check back in a while and see where it went. I'm working on an "adapter" (so-to-speak) that queries SQL, so maybe I'll add yours too when the time is right.

erezsh··on Show HN: CEO, Decision Making as a Service
Except you have very little context, and you get all of it from a person who's desperately incapable of deciding.
erezsh··on SQLite Is Serverless
Yes, that's what I meant. And possibly for ML preprocessing.

Just out of curiosity, who do you imagine your users will be?

erezsh··on SQLite Is Serverless
If you have full support for SQL, it means some queries can run for a few seconds, possibly even minutes (even sqlite supports some form of bfs). So will you just time-out those requests? That would make a lot of possible uses suddenly impossible.

I might consider the unmetered option, if the performance gain is worth the 6x cost of other providers.

For web and games, maybe your existing model could work, but then always having a recent backup also becomes more important (and for most sites, the lack of acid is a deal-breaker).

The nice thing about data-science is that you interact with the workspace yourself, so you know exactly when you want to save a snapshot.

erezsh··on SQLite Is Serverless
I'm far from an expert on such things, but I like the adjustable fixed-price model (like digital ocean).

Whenever I see a price like 0.0002$ per hour, I feel like they're trying to mess with my intuition.

That's especially applicable to data-science, because you're not worried so much about automatic scaling, you just don't want to be surprised at the end of the month.

I don't know if you can match digital ocean's prices (they offer managed postgres, for a fair comparison), but if you can get close then you have a chance.

erezsh··on SQLite Is Serverless
Looks pretty cool!

I can imagine it would be useful for data-science projects, if the pricing is right.

erezsh··on New study: The advertising industry is systematically breaking EU law
I'm still "waiting" for ads that are actually relevant for me, and not just based on something I randomly watched a few days ago.
erezsh··on What Is the Most Valuable Thing You Can Learn in One Hour?
I've been using Vim and vim-derivatives almost exclusively for the past 15 years, and I think none of the things you mentioned are important to use vim.

The first and most important things to understand are:

- Normal mode is the default mode. Only leave it when you already know what you want to write.

- Movement follows the [<count>] <verb> <target> model. The most used verbs are: d, y, p

The most used targets are j, k, f, F, /, w, W, e, b

Other useful keys to know and use: a, A, o, O, *, ., %

And lastly, learn how to use macros! (q, @)

If you just learn everything I listed here (and it isn't much), until it feels natural, you already know 90% of what you need to boost your editing speed and use vim like a pro.

erezsh··on WTFPython – Understanding Python through surprising snippets
This list is a little nit-picky.

All the id() equivalencies only exist for optimization, and not out of some promise to the user.

The walrus operator can be confusing, but any language can be made to look confusing. It's an advanced feature that should be used very scarcely (if at all).

It does point out a few real weak points in Python, but most of them require non-trivial usage of the language.

Meanwhile, in Javascript

    > "a" + undefined
    "aundefined"
Now that's what real WTF looks like.
erezsh··on Glush: A robust parser compiler built using non-deterministic automatons
That sounds really cool, and I hope it will find success. The field of parsing truly needs some refreshment, especially in usability.

I do have a few questions:

1. You claim the algorithm used guarantees O(n^3) for all CFGs, but is different from Earley and CYK. Are there more details for this algorithm, perhaps with a rigorous mathematical proof for this claim?

2. How is Glush's performance in relation to Earley, in terms of constant time and memory requirements? (even for low complexity, a high constant can be problematic)

3. How do you deal with ambiguity? The blog mentions priority, but is there a way to get something similar to a SPPF?

erezsh··on Shopping Sucks Now
To be fair, have you met people?
erezsh··on Show HN: Assembly: A Pythonic Object-Oriented Web Framework Built on Flask
From the tutorial:

    class Index(Assembly):
      index(self):
        return 
How is this valid Python?
erezsh··on The Efficiency-Destroying Magic of Tidying Up
> Streets arranged in grids, people waiting in clean lines, cars running at the same speed…

Grids make it easier to navigate for humans, to plan for public transport, and improves the city's connectivity.

Waiting in clean lines reduces the chances of fighting in the queue.

Cars running at the same speed dramatically reduces the chances of a lethal impact.

And what exactly is the cost of these things? Is a grid that much harder to make? Does anyone want to cancel the dedicated highway, or let cars stop in the middle of it?

I think this author just doesn't understand the concept of efficiency at all.

> “Please clean up your room,” asks the mother. “Fool,” retorts the three-year-old with an eerily deep voice. “Can’t you see the beauty in my glorious chaos?”

Sounds like he's still upset about that..

erezsh··on Introducing nushell
Maybe you're just not very imaginative. I can see my workflow improve considerably by using a structured-io shell, and I'm already quite proficient in bash. So, I'm very happy some people are trying to make it happen.

Also, I'm not sure what your argument is, except "I don't get it"? It's fine, lots of people don't get stuff. Just move on, and maybe in a few years when it's mature, you'll see it again and go "Ah!"

erezsh··on The Rascal Metaprogramming Language
How is GLR more general than CFGs?
← PreviousPage 6 of 8Next →