olavgg

For what I am currently doing, years, if not a decade or maybe even two. Claude has already been running for 3 months constantly working on this project. I've been trying to do this several times before, but the scope outgrew my available time/life, too big for one person. But now I can actually build a production release by just setting the direction and let Claude to all the wiring, plumbing, tests etc.

I do now add all the major features I want, and write a task list where I add a few more minor tasks each week that just come to my mind. I need to write them down before I forget. I often ask Claude to implement several different implementations/algorithms and compare them. There are plenty of trade-offs, which usually end up in building configuration profiles, so the application can work optimally in multiple environments.

The specification is actually really large, and has grown over time as I've been learning and new ideas has come to my mind. I do not review/audit the code, it's massive would take me months, no chance! I do however use code inspection/audit tools like TSAN, ASAN and more. I will also spend time on full code reviews with the coming Fable 5.5 when a release is about ready. I do however review behavior together with Claude, and this is also where the specification grow, because this and that should be possible, but not this and that. User stories also helps.

Yeah, I should see if I can create a similar skill. The tests do spend some time to complete.

Claude: Confirmed: 195M of the 212M cache tokens (92%) come from full re-writes after 5+ minute gaps, accounting for 77% of relative cost. Now I'll check the existing wrappers so the skill builds on them.

Its easy when you spawn a couple of subagents in a homelab with a rack with 5+ servers.

This is how one of my claude code session looks like right now, one prompt from yesterday:

  ◯ lean-tracks  ▰▰▰▰▰▰▰▰▰▰▰▱▱▱▱▱▱▱▱▱  7/12 · 18h35m · ↓ 3.1m tokens
  ◯ lean-tracks  ▰▰▰▰▰▰▰▰▰▰▰▰▱▱▱▱▱▱▱▱  6/10 · 17h47m · ↓ 2.6m tokens
  ◯ lean-tracks  ▱▱▱▱▱▱▱▱▱▱▱▱▱▱▱▱▱▱▱▱  0/1 · 4h49m · ↓ 275.9k tokens
  ◯ lean-tracks  ▱▱▱▱▱▱▱▱▱▱▱▱▱▱▱▱▱▱▱▱  0/1 · 4h30m · ↓ 395.0k tokens
 ◯ lean-tracks  ▱▱▱▱▱▱▱▱▱▱▱▱▱▱▱▱▱▱▱▱  0/1 · 3h50m · ↓ 490.3k tokens

I really like the having DeepSeek Flash + Opus 5.5 as advisor/reviewer at the moment. I run through a $200 Claude Code subscription in less than 24 hours now just for a single project, and I hate fiddling with different accounts every day. Now that I get a 200 USD api credits inlcuded with the subscriptions I have rewritten my coding agent workflow completely.

Shameless plug...

MatterMost was actually a great file sharing platform for our business. But they was not true open-source and removed features we needed like SSO.

So instead we vibe-coded a MatterMost Slack alternative which we have been running on the internet fine for months now https://intellistream-datahub.github.io/intellistream-chat/ Its super convenient to share files on. And it comes with built-in secret sharing like YoPass. We occasionally implement new features and fix bugs we discover.

Its one jar file, and supports Vault/OpenBao and requires Keycloak and PostgreSQL. Lucene is embedded.

I do the same, a single app on a zfs filesystem I can snapshot/send/receive

I play BAR a few hours every week. Online games are really good and smooth. BAR is by far the best RTS out there today. It only misses an awesome single player, but its coming now.

Some of the elements in BAR is really awesome, radar, anti-radar, radar killing rockets, EMP on anti-nuke. The strategic depth and complexity is really at DOTA level.

It's by far the greatest RTS to play for the next 10+ years. And its free.

Pulsar 5.0 has two headlines. Scalable Topics are a new kind of topic that grows and shrinks with demand while preserving per-key ordering. The second is major performance and reliability improvements for the topics and applications you already run, which you get by upgrading your cluster, with no application changes. Oxia is the recommended metadata store for new clusters that replaces ZooKeeper.

The numbers are quite interesting, does bcachefs run on Almalinux 10? I only see Fedora supported. All my servers run Almalinux today and I am super happy with it, no plans to use something else.

How is root support for bcachefs? This is the one thing I really miss with ZFS today, its just too much work. I do love my FreeBSD systems with a root ZFS though.

I have already created a croos-platform solitaire game in Rust with Opus 4.5. Done in one evening. I absolutely do not understand why OpenAI could not write a proper application in 6 months. Do they struggle to use the product they sell?

ROCE do require proper nic/switch tuning. But if you go all in on Mellanox NIC's and switches, its basically on knob on the switch: roce

If you buy CX3 and Mellanox SX6036(easy fan adjustment and low power usage) switches on Ebay for cheap. Which are great up to 56Gbps. You can run infiniband which is way easier to setup, just remember you still need to enable IP over IB to start the handshake for the RDMA connection.

Yes, nfs with rdma will increase your performance significantly.

I fully agree with the author, and I have stayed on vanilla JS for decades now. I like the simplicity, ownership, zero dependecies. I have my own bundler in Spring Boot using Google Closure Compiler, which works autonomously with the javascript assets. No nodejs needed.

Also with the developer disruption of agentic coding AI, the arguments for using vanilla JS is even stronger. Code reviews of code/solutions is much easier. Scoping of code paths for each server-side render templates is easier. The AI does all the boilerplate coding, while keeping it readable and debugable. Implementing design patterns is just common computer science stuff, and the coding agent is pretty good at it.

I also what to mention, as a business owner, and boss of a startup. All developers here work with vanilla javascript. They were first surprised, but today they prefer it by a huge margin because of the simplicity and zero dependency and non existing build steps. The web browser is the framework, you need another abstraction over the web browser.

The MSSQL developer experience is ok, until you want to pipe the backup over ssh. Then you understand that the Linux support is a joke, with no stdout(https://linux.die.net/man/3/stdout) support.

Yes CX5 can offload more. I believe CX4 has similar offloading capabilities as CX3, except that it supports 100G.

Another note: In my experience, RoCE works much better on CX4+ generation. CX3 is best with Infiniband. I think some firmwares on the CX3 generation, has a messed up config for RoCE. But running Infiniband is not a complex task, is way easier than people think, like 10x easier and faster to setup than Ethernet.

Maybe you have priced it wrong? I just checked Ebay, a 16GB 12800 Registered ECC module goes for 40-50USD ea. That is crazy! Last year they were like 5 USD each.

With NVMe-oF and RDMA I can saturate a 25G link with spinning disks easiliy with around 16 drives.

I disagree, I play a lot around with enterprise stuff. Its insane that I need to buy enterprise grade hardware that costs 1000x more for lab/experimentation/learning. My only alternative is to wait a few years, and get it from Ebay.

I also believe that a strong reason that Optane pdimm's failed, was that it was only available on enterprise servers so hackers didn't get a chance to play with it and build software that took advantage of this special hardware.

Just look at how specialized Infiniband is, even though its awesome and has some great use cases. If it was a commodity tech, there would be 100x times more applications/software that took advantage of it.

We're a self hosted shop, we went with Valkey. Valkey also has support for RDMA, which we already is running in our infrastructure.

But that is his point. If you cannot find the session id in redis, you login again. If your Redis server crash, you start a new one and everyone just login again. No data is lost.

When I look at the oil & gas sector, I remember MS-DOS + Wordperfect was the beginning. Then Windows 3.1 + Microsoft Office took over, and since that, its been Microsoft, Azure, and SAP.

They refuse Google Cloud, AWS, and many still believe open-source is cancer. They are Microsofts best customers. They prefer consultants over hiring software developers, and the consultants just to what they're told and never question the status quo.

Whenever I spending time at these companies, my head is filled with dinosaurs.

Where I live we have something called The ONS event/Exhibition, where the oil sector gathers to promote themself. 2 years ago AWS had a big stand there, but it was mostly empty. This year, AWS doesn't participate at all.

I just paid 120 000 USD for replacing the exterior drainage system. I knew it was expensive, but not that expensive.

Every time I give Photoshop a chance I give completely up, the thing doesn't even start on neither Fedora or Ubuntu and I have no interest in configuring Wine for this. GIMP is the least painful way to get the job done. I have been using it for over 20 years, and it has been a pain-free experience. That GIMP is bad is just as wrong as the people who say Java is bad.

But you know what's even worse, people that use Illustrator to create SVG's for the web. Inkscape creates proper readable SVG's at 5KB, compared to 50MB SVG's I get from Illustrator experts.

It is a bit different, Mellanox with iSCSI/iSER over IB/RoCE is much less complex than iSCSI over TCP. RoCE runs over UDP, but requires switches with PFC, and ECN. Chelsio plays nice with any switch. Mellanox has much better offloading of NVMe-oF, where the network card can directly communicate with the NVMe device over pci-express so it can completely bypass system ram and host cpu. In fact, Linux will know nothing about the transfers to and from the NVMe device.

To get accelerated iSCSI, you need to install Chelsio or Mellanox drivers. While both work out of the box with the inbox drivers, they have special drivers that you have to download and install to unlock the extra performance. I think Chelsio has everything included in FreeBSD, so there the inbox drivers already come with top performance unlocked.

ZFS supports self healing, you do not have scrub, it will be corrected during a bad read as long you have a copy. Metadata has 2 copies by default for additional safety for a single disk.

Using a proper NIC (Chelsio) with their iSCSI accelerator will boost your iSCSI performance significantly. Another alternative is Mellanox with RDMA. You need CX4+ for optimal performance over TCP/IP, while the cheap CX3 is excellent with IPoIB. If you have a lot of packet drops and retransmissions, another option for boosting iSCSI performance is getting a network switch with a lot of memory for packet buffering. This helps with incast congestion. There are special switches with gigabytes of memory built for this.

NVMe-oF is the best protocol with least overhead for network drives, with a proper setup you lose only 10-20% latency compared to local disk even with Intel Optane. Throughput should be almost similar.

A 42U rack filled with 1u servers with 8 drives each, will have 84PB of data. It feels like it was a few month ago where you could buy a rack with 1PB of storage, and that was awesome. Not anymore.

I would not worry about running a distro. Most things are similar, there are some minor differences between the GNU applications and the FreeBSD alternatives. But for most people there is nothing to worry about. Most applications runs fine on BSD. Bind, PowerDNS, Java Applications, PostgreSQL, Python, rsync and many more. Getting PyTorch to work with Nvidia and Cuda is most likely another story.

My main culprit with FreeBSD is that upgrading the kernel is not a simple dnf update command. But its still easier than upgrading RHEL from 9 to 10.

I just started with FreeCAD this weekend, and in 3 hours I managed to create a simple pci bracket that I could 3dprint. I just followed some YouTube tutorials. When I learned 3D Studio Max 25 years ago, I struggled a lot more.