HNHacker News
TopNewBestAskShowJobs

amirouche

929 karma · joined September 20, 2012

"CC" [ @1 ] APPLY
submissionscomments
amirouche··on Build Impossible Programs
> The 80/20 Rule seems to work quite well here, when I see it applied, which is not often enough.

That is 20% of the effort can deliver 80% of the features?

amirouche··on Build Impossible Programs
At last someone speak up about that very subject "doing impossible things" without being negative about the fact the community (of haters) forbids that behavior.

Innovation is NIH somewhat, somehow and that is a Good Thing.

amirouche··on Reazon – miniKanren for Emacs
I think nothing can beat minikanren (or microkanren) to query a quad store. It make recursive queries easy.
amirouche··on How I Built a Personal Knowledge Base for Myself
> searching is far more effective than organizing. My knowledge base became a buffer for my web search.

Searching doesn't with keeping a copy of the thing you need around.

amirouche··on LiteTree: SQLite with Branches
Uses cases for a versioned in branches data store that are not blockchain related.

0) The most obvious use case is to replace git, because git proper doesn't handle big files very well, let alone bigger than RAM structured data. Mind the fact you might still need an ad-hoc solution for binary data. WT does have a limitation on value size.

1) There is various data science uses cases, like sharing data [0] or A/B testing stuff.

2) Anywhere you need an audit trail .ie you need to look back at previous states, whether it is for debugging purpose or domain logic

versioned datastore is a compromise between event sourcing (pure log) and non-versioned databases.

[0] Collaborative Open Data versioning: A Pragmatic Approach Using Linked Data

[1] https://en.wikipedia.org/wiki/Audit_trail

edit: cosmit.

amirouche··on LiteTree: SQLite with Branches
Why did you choose LMDB among leveldb, wiredtiger and bsddb or even gdbm?

It seems like you do not rely on range queries at all.

amirouche··on LiteTree: SQLite with Branches
> LiteTree is implemented storing the SQLite db pages on LMDB.

Why are you doing it like that? Does it lead to some limitation of some sort? Like making merge very costly?

amirouche··on LiteTree: SQLite with Branches
That is up to the upper layers to decide which record wins.

In my implementation, the default is to take the records from the "other" branch because you want 'devel' in 'master' not the other way around. Not sure I am clear. LMK.

amirouche··on Ask HN: Has anyone built a recommendation engine in-house?
Same situation with a smaller project. I still struggle to pull off it for the problem I face today. How did you succeed?
amirouche··on Python’s Meta-Object Protocol (2012) [pdf]
LISP languages do this better than python.

edit: plural

amirouche··on Ask HN: What was the first side/hobby project you completed?
It's never finished. 1) Some people invest time almost a single code bases 2) others start and stop numerous projetcs 3) others do nothing. And everything in between.

Myself, I am in the second category. I start but not maintain project.

Here is a list of feature complete repositories:

- https://github.com/amirouche/scheme-todomvc - https://github.com/amirouche/deuspy - http://quicoeur.fr/ - http://sensimark.com

I have other projects like:

- https://framagit.org/a-guile-mind/culturia.next/

- https://framagit.org/a-guile-mind/culturia.one/

- https://github.com/amirouche/zehefyu93

- https://github.com/amirouche/socialite

They are not done because they are more research work on various topics like UI/UX that is more like CLI in the browser this lead to the development of a chatbot like experience in the spirit of ipython. In the database side, I think I will recode culturia.next. Even if I don't know what the future will bring.

Worse, I have two go to stack:

1) Guile + BiwaScheme or simple SXML

2) aiohttp + reactjs like socialite maybe i will move to https://github.com/amirouche/beyondjs/ with https://github.com/amirouche/found

amirouche··on How NoSQL forced the evolution of a scalable relational database
Why EAV is plague? Isn't JSON type a wrapper of EAV?
amirouche··on Goodbye Microservices: From 100s of problem children to 1 superstar
what about idempotent updates?
amirouche··on Goodbye Microservices: From 100s of problem children to 1 superstar
> it's much easier to process the test output and debug the microservices than monolithic applications.

You easier to debug end-to-end tests of a microservice architecture that monolith? That's not my experience. How do you manage to put side by side all the events when they are in dozen of files?

amirouche··on Important database news from the last few months
How is realistic this article without a word about FoundationDB?
amirouche··on Getting Started in Natural Language Processing
CVPR?
amirouche··on Show HN: Zehefyu93 Scheme^2 Quad Store for collaborative structured data
Here is a screen cast showing some features https://www.youtube.com/watch?v=YMYh6f_jODg&t=6s
amirouche··on Event Sourcing made Simple
I agree that SSD storage is not cheap. I think that you can offload 'old' data to S3 or something.
amirouche··on Fabric 2 is out
I like the fact that fabric 2 is a library first, instead of somekind of 'recipe' language. Putting aside invoke in its own library is a great move. Now, I can happily avoid Ansible.
amirouche··on Ask HN: What tool do you use to do advanced custom graphs in?
inkscape?
amirouche··on Everything old is new again: Microservices
I read the whole topic looking for enlightenment and insights to understand micro-services. I've been only in small companies (less that 10 years old with a dozen of engineers each) and AFAICT services grew organically without much architecture supervision or even if there was architecture supervision, it was a way to escape the technical debt of previous services. Don't even add to that the technical difficulty to manage multiple services in development or production or in between.

On all the code bases I saw, I never told myself (micro) services are helpful, I will redo that. And if I had to rewrite any of them, I would use a single 'service' that can do everything. IMHO it's more sane to optimize for the developer instead of whatever micro-services try to optimize.

amirouche··on Everything old is new again: Microservices
How a network protocol is more specified that a procedure call?
amirouche··on Everything old is new again: Microservices
Yeah, but aren't Erlang project a single code base?
amirouche··on Everything old is new again: Microservices
What do you mean by 'database instance'? Is the the instance of the 'class' that represents the database and its records or something else?
amirouche··on Ask HN: How do you memorize syntax and common API for a new language?
I know sexp; I don't need to learn new syntax. I only need to learn new API.
amirouche··on GUI development is broken
2. is fixed by termbox
amirouche··on GUI development is broken
What about termbox, it's a good enough abstraction for coding application for the terminal.
amirouche··on 81 Year Old Commodore Amiga Artist – Samia Halaby [video]
+1

(Sorry HN etiquette)

amirouche··on What I learnt from building 3 high traffic web apps on an embedded kv store
What is Lustre?
amirouche··on What I learnt from building 3 high traffic web apps on an embedded kv store
The monolitic nature of embedded key-value stores based webapps makes it a poor solution for scaling horizontally. But who cares?

Mind the fact, that given the database is embedded in the webapp processus, you need some auth machinery to hit the database from the outside world anytime. Which leads to more wheel reinvention like background process scheduling like Celery.

At the end of the day, embedded key-value stores are great because they require _less_ administration and have great performances but the community is fragmented and tooling is not there yet.

I think embedded KV stores have still a chance in niche applications for building expert systems that require persistence but don't change a lot. For a web application which specifications (and schema) change all the time, it's a poor solution.

edit: also, in the context of e.g. wiredtiger having transactions across tables like the full-text search index and plain tables is very great and save quiet a few painful experience of setting up the project.

← PreviousPage 4 of 17Next →