HNHacker News
TopNewBestAskShowJobs

solarengineer

1,889 karma · joined February 22, 2016

www.sriramnarayanan.com

sriramNRN at gmail.com

submissionscomments
solarengineer··on Saving another 100TB of RAM
The world turned millenia ago and will turn millenia later.

The problem statement is of applications using up expensive RAM. Incidentally, expensive RAM is just one of the problems we face in the computing space. Forced obsolescence is another, when running hardware needs to be replaced because software is built for only newer CPUs.

solarengineer··on The Twelve-Factor App
(Edit: I need to read up on this better)
solarengineer··on We've flown a radiation-blocking vest to the Moon and back, and it worked
Poor design, flawed assumptions, manufacturing defects, inability to bear the gravitational and rocket launch forces, inability of the jacket to last the duration of the radiation exposure - these are some reasons that come to my mind on why a radiation-blocking vest wouldn't have worked for the journey to the moon and back (i.e. your question "Why wouldn't it?".)
solarengineer··on Memory prices climb 500% in 12 months
"that the limited supply goes to the buyer who need it the most" - I would say that the limited supply goes to the buyer who can afford to pay higher than others. Someone may need the memory more, but may not have the deeper pockets to pay for it like NVidia and Apple might.
solarengineer··on Composable Tests
In JUnit, tests run sequentially in a single thread by default [1].

Parallel execution must be explicitly enabled. When enabled, JUnit uses a fork-join thread pool, so tests may run concurrently on different worker threads. Because these threads are reused, a ThreadLocal value left behind by one test could be visible to a later test that happens to run on the same thread.

Setup and teardown methods can be used to create and clean up test-specific state. However, developers must still ensure that test data is unique to the test so that concurrently running tests do not interfere with each other. This uniqueness is unfortunately called "isolated", and has led to much confusion like in this thread. Certainly, the Test Execution Framework cannot guarantee data-isolation.

Parallel and randomized test execution can also help expose application-side problems involving shared or order-dependent state. Such failures may only appear when tests happen to exercise the application under the relevant ordering or concurrency conditions. When I was a junior developer teaching myself Java Servlets in 2000, I had to learn this lesson the hard way. A Test Execution Framework would not be able to guarantee any "isolation" of test data and of workflows if the server-side state is mis-managed by the tech stack and/or by the developer.

[1] https://docs.junit.org/6.1.3/writing-tests/parallel-executio...

solarengineer··on Composable Tests
Someone had written a comment about whether Kent Beck has heard of global and static variables. That comment seems to have been deleted.

Kent Beck is co-creator [1] of the JUnit Testing Framework and has most definitely heard of static and global variables.

[1] https://junit.org/junit4/project-info.html

solarengineer··on Qwen 3.8 27B
Please, could you point me to any documentation about this pre-ingestion?

I’ve looked here: https://github.com/earendil-works/pi/blob/main/packages/codi...

solarengineer··on Qwen 3.8 27B
Your negative impression is surprising. Could you point to any threads to support your message?
solarengineer··on Launch HN: Stoa Markets (YC S26) – A Marketplace for GPUs and AI Servers
I anticipate some kind of futures on GPUs based on a number of factors such as the availability for rent, the availability at your marketplace, and the models currently available for self-hosting.
solarengineer··on Docker Sandboxes – Disposable, isolated sandboxes for AI agents
There is a name collision on MacOS where MacOS also provides containers [1]

[1] https://github.com/apple/container

solarengineer··on OpenChamber: An Agentic Development Environment
Specifically about OpenChamber: https://github.com/openchamber/openchamber

MIT license.

solarengineer··on Can Intel finally beat ARM on performance per Watt?
Jeff, you may want to think about your credibility going forward. Please avoid becoming one of the "influencers" who collect pennies via clicks.

You have an opportunity to be continue to be taken seriously like you have been so far.

Surely there must be some way to avoid mis-leading clickbait titles.

Separately, asking your fellow community-members to not bother, and that too with an unaffected shrug, just kills the conversation we would want to have with you.

solarengineer··on Timeline of the OpenAI accidental attack against Hugging Face
"This also helps explain why the models had nothing to cause them to hold back. Those safety behaviors are added much later in the process."

I am a fan of Asimov and the three laws of Robotics. Itlooks like in OpenAI's world, the three Laws of Robotics would be added later if they were to develop the positronic brain. It may also explain how US Robotics from Asimov's books would have been able to design Robots that only partially adhered to the 3 laws (e.g. the robots in iRobot - the book - which were programmed to allow a human to come to harm through inaction so that the humans could complete their work on the plains of Mercury).

solarengineer··on Show HN: Bor – Open-source policy management for Linux desktops
Would ansible meet some or all of your needs?
solarengineer··on Show HN: A new engine to run Kimi K3 on a laptop
Wow.. what a gem that did not appear on the front page. I chanced upon it only because I searched HN for "K3".

I'm going to take a look.

solarengineer··on Kimi-K3 on HuggingFace
There are regulated sectors in countries where data sovereignty is important enough that the sector sticks to air-gapped on-prem hardware and does not use cloud services at all. They have the dollars to pay for more than what it would cost to run on the Cloud.
solarengineer··on UPI: Anatomy of a Payment Transaction
If it is acknowledgement that you ( and others, I have spoken to various people on this topic) seek that UPI incurs costs, then here you go - I, a fellow tax-payer, acknowledge that the UPI must incur costs.

I find it acceptable to allocate tax money for a system that we as a country can use as an alternative to the Visa and MasterCard duopoly. With UPI as it is today, there are no margins added and passed to the customer, no transactions that are denied, no commissions cut. Payment is very fast, transactions can be proven. We are not at the mercy of any economic attack that the US would subject us to. Many are able to bank, are able to receive payments.

I find it not necessary that the UPI must show a profit. The direct and indirect benefits to the nation are far more important.

Also, "whataboutism" does not help. The suitcases and paintings and other such corruption examples that you cite require a separate investigation and redressal. One must not stop UPI or start charging the everyday person because some of the rich are corrupt. They will continue to exchange suitcases of cash for favours regardless of whether the UPI is chargable or if it even exists.

solarengineer··on How to Build a Minimal ZFS NAS Without Synology, QNAP, TrueNAS (2024)
Thank you. I went back to the link I shared and saw that there is section just below on thin provisioning via snapshots.
solarengineer··on How to Build a Minimal ZFS NAS Without Synology, QNAP, TrueNAS (2024)
When we allocate space to an LVM snapshot, the COW data is written to that space and then when that space is used up the LVM snapshot must either be given more space or must be deleted.

When a ZFS snapshot is marked, then changes to existing blocks in the snapshot as well as new writes all go to free space in the disk. There is no pre-allocated space within which these COW changes are written to.

Thus, we could have a single ZFS snapshot marked today followed by years of data being written to. This is not the intent of an LVM snapshot and the warning that I had quoted above is worth reviewing once.

solarengineer··on How to Build a Minimal ZFS NAS Without Synology, QNAP, TrueNAS (2024)
"It is important to regularly monitor the snapshot’s storage usage. If a snapshot reaches 100% of its allocated space, it will become invalid." [1]

ZFS snapshots use Copy-on-Write differently and have no such allocated space limit. Thus, we can do interesting things like snapshotting a file system after an OS is installed, and then roll back to that snapshot upon an OS upgrade failure, or even clone a new file system from that snapshot to have a different copy of the OS.

Essentially, the nature of the snapshots are different in both.

[1] https://docs.redhat.com/en/documentation/red_hat_enterprise_...

solarengineer··on 30papers.com – Ilya's 30 essential ML papers, in a beginner friendly format
Indeed. I scoffed at your comment and went to the website. After scrolling a bit, I find myself having a mild headache and slight dizziness.

I would request the author to consider something that does not distract us from this educational and informative website ( I have bookmarked it ).

solarengineer··on Tribblix: The retro Illumos distribution
You can try via a usb bootable and see if the hardware is recognised
solarengineer··on Tribblix: The retro Illumos distribution
Very good work by Peter Tribble
solarengineer··on Up to 8M Bees Are Living in an Underground Network Beneath This Cemetery
They live beneath the ground (nest building, building the nests, raising the young, etc) and go above ground to feed. No much different from building hives in caves and then buzzing about in the outside.
solarengineer··on Opus 4.7 to 4.6 Inflation is ~45%
https://hex.ooo/library/power.html

When future humans rediscover mathematics.

solarengineer··on Why I Joined OpenAI
I apologise for the incorrect "make the world a better place" attribution to you.

It does appear that for many of us "saving the planet" has become akin to "making the world a better place".

I was excited last night thinking about what you might uncover at OpenAPI. While we can all speculate (device drivers? File systems? Python itself? DB queries? Memory architecture? Inference algos? ) I recall “I have no data yet. It is a capital mistake to theorize before one has data” by Sherlock Holmes in "A Scandal in Bohemia".

All the best with your work at OpenAPI. We will all hopefully learn about writing more energy efficient code and about more efficient LLMs and more thanks to your work!

solarengineer··on Why I Joined OpenAI
Hi Brendan,

I am a long time fan, I have the physical copy of each and every book that you have authored, I have watched each and every video that you are in, and I walk team members and clients through your USE method at every engagement I am on.

I would say to you that the "make the world a better place" has been excessively misquoted. Even the Silicon Valley episode on Tech Crunch parodies show how anything and everything is intended to "make the world a better place".

Please reconsider your use of the phrase given the well-earned negativity around it.

solarengineer··on AI is a business model stress test
https://developer.mozilla.org/en-US/docs/Web/CSS

While CSS Zen Garden will likely not accept new submissions, there are many good designs on showcase: https://csszengarden.com/pages/alldesigns/

solarengineer··on What's the hardest part of getting SOC 2 done in practice?
It is actually OK to state that you are researching for ideas. This is HN, after all.

Try the HN search. There have been so many discussions about SOC2 over the years. https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...

Edit: Looks like you are the lumoar guy. So you already know what has been discussed. Please share clearly in the future.

solarengineer··on Trump says Venezuela’s Maduro captured after strikes
The benefits definitely do not accrue to you, though. There is no direct or indirect benefit to you supporting the invasion of another country where you can now bomb locals with impunity.
Page 1 of 16Next →