inapis

With tailscale, you can establish a private LAN over the internet. Iroh embeds this directly into the app itself.

Use tag-based node authentication. Login as a user and then switch the device to use a tag. I just recently did that and retained the usual 6 months expiry. I can also disable key expiry completely.

Skill issue. Not to mention the ongoing effort required to maintain and secure the service. But even before that, a lot of people are behing CGNAT. Tailscale makes punching a hole through that very easy. Otherwise you have to run your own relay server somewhere in the cloud.

What? You still have to install the binary via your package manager. Most OMZ plugins are basically a bunch of shell completions or aliases.

Yet if I spent one hour making my app one second faster for my million users, I can save 277 user hour per year. But since user hours are an externality, such optimization never gets done.

I have never been convinced by this argument. The aggregate number sounds fantastic but I don't believe that any meaningful work can be done by each user saving 1 second. That 1 second (and more) can simply be taken by me trying to stretch my body out.

OTOH, if the argument is to make software smaller, I can get behind that since it will simply lead to more efficient usage of existing resources and thus reduce the environmental impact.

But we live in a capitalist world and there needs to be external pressure for change to occur. The current RAM shortage, if it lasts, might be one of them. Otherwise, we're only day dreaming for a utopia.

I am not a system programmer but, from my understanding, Torvalds has expressed strong opinions about microkernels over a long period of time. The concept looks cleaner on paper but the complexity simply outweighs all the potential benefits. The debate, from what I have followed, expressed similar themes as monolithic vs microservices in the wider software development arena.

Sure. Till an extent. And if you run some mission-critical application, definitely.

But most applications run fine from local storage and can tolerate some downtime. They might even benefit from the improved performance. You can also fix the durability and disaster recovery concerns by setting up on RAID/ZFS and maintaining proper backups.

No. DO can be equally noisy but I've always tried their regular instances and not their premium AMD/Intel ones.

Inability to build a business easily nor allow others to build one. The union culture is way too strong. There is a joke that the Kerala model of development requires a rich oil state nearby.

As someone who has struggled with this before, your fastest way is to look for pre-built templates on github or buy one's you like.

Looks like GP is using the pocket template from tailwindui - https://pocket.tailwindui.com/

There are hundreds of people selling pre-built components, landing pages, templates etc. This at least gets you up and running and not stressing over design. As a dev who lacks design sense, this was immensely helpful.

Other option as sibling points out is to use bolt or lovable and give it explicit instructions on what kind of design to use. For example, with lovable, try this prompt "use neo brutalist design."

If you don't have a lot of services to access, you can hard code the tailscale IP address in /etc/hosts.

My personal /etc/hosts is at 10 services all hard coded since the internal IP address of a machine on tailscale is static. Way cheaper and easier to deal with than setting up a separate DNS resolver.

Of course that won't work if you have hundreds or thousands of services to work with.

If I may ask which model are you using? I have tried OCR'ing my bank statements in AI studio and the results have been less than optimal. Specifically it has a tendency to ignore certain instructions combined with screwing up the order.

Some pointers on what worked for you would be greatly appreciated.

don't ask them if they want it or not

After reading the book my understanding was to not even ask this question. See if they have already tried to solve the problem by looking for another app or going to excruciating lengths to solve it another way. If yes, then they will probably buy your product.

Jaypee Group's failure is not the same as ghost cities of China. Ghost cities of China were fully built but people refused to move in. India has the opposite problem. Most people want to move in to the projects but the builder defaulted and could not complete the project.

India's half built cities are more of a financial fraud problem rather than people problem.

Not the same as China.

Nowhere did I imply that.

Anecdotal experience but the bike looks ugly + had a subscription at one point in time (not sure if true today) for accessing basic features.

so in a quick glance, this does not look like nginx/caddy. It is not a binary you download, install and just configure to talk to your upstream servers. Rather a set of packages which you assemble to fulfill a particular use case. You basically end up writing a "new" reverse proxy just for your use case.

Not sure how useful this would be for anyone except very large businesses or someone like cloudflare itself.

I don't think there's anything, besides economics, preventing current technology to achieve the same reliability.

No one can say for sure how the entire thing is going to turn out. As investors, there's no harm in placing multiple bets as long as a small number of those bets end up giving an outsized return.

Not really true. ADATA pretty much demonstrated that you could get away with swapping the controller at least 3 times and slowly shaving off performance so that people won't notice until they actually did. And the performance was down by 25%-40% in the third swap. This sets a precedent and sooner or later other manufacturers will follow.

My pet peeve with the whole debacle is that words and promises no longer matter especially when marketing ends up using words like "upto."

Give me a minimum, maximum and specs written in stone and I won't care if you change out components. As it is now, the entire thing has become a game of whack-a-mole, SSD reviews are loosing their relevance and given enough time (+current trajectory) we could easily end up in SSD market which is more like fast fashion than anything else.

Eh. Governments should absolutely not get in the business of creating or running an app store. Mobile computing is, effectively, a necessity in the modern world but the web already serves as a great independent distribution platform. We should rather focus on not enabling the technology giants to gimp the web for their business moats.

Fair point but I am not sure how it is any different from other risky industries. Aviation became safer (current Boeing fiascos not withstanding) over the past 2-3 decades where today an airliner crash makes global news.

Humans being terrible at something, at some time in the past, is not a sole reason to write off the technology. The response should be to rather focus on improving the process and policy rather than shrugging off nuclear which IPCC has admitted is necessary to meet our climate goals.

Most American's are very worried about dying from terrorism while most are actually being killed by heart attacks.

At a certain point it is a matter of what people want to believe and what is actually happening on the ground.

There hasn't ever been a nuclear plant built which as not subsidized. Wind and solar needed a kick-start, now they bid to even get permission to build off-shore wind power.

I am not sure why nuclear needs to be perfectly competitive to be a valuable option. Govts worldwide subsidize infrastructure programs to varying degrees. Roads are subsidized. Water treatment plants are subsidized. Hell, agriculture and farming cannot survive a free market without government subsidies in most of the planet. Companies like Apple and Tesla won't build manufacturing plants without tax breaks (effectively a subsidy.) Amazon won't build an office without asking for a tax break from the cities.

The expectations of nuclear are irrationally high.

And your quotes all point to policy issues which can be fixed with proper governance. Just because it "became" non-competitive 5 decades ago is no reason to not pursue it going forward given that any help against the climate emergency is welcome. Even IPCC in a report some time back accepted that meeting the climate target is not possible without the help of nuclear and in a good policy environment nuclear can be cost competitive.

Really the whole debate is mostly scientism: nuclear power didn't lose because it was too dangerous. It lost because it is too expensive. These days, it's still too expensive, but now it also just takes too long for it to be useful for the climate emergency.

It's expensive because Chernobyl caused the whole world to develop cold feet. Innovation stopped and funding for nuclear dried up after 2-3 accidents. Solar and wind were super expensive (vis-a-vis coal or gas) in 2010 and before compared to 2022 and people were railing against solar for that reason alone. But only a decade later, the cost has fallen by an order of a magnitude, making it far more practical and a valuable source of energy because society invested money and resources to bring the cost down.

Imagine having 3 children and lavishing all your attention and resources on two of them, leaving one to fend for itself and then turning around and saying that the aloof one has no talent, skills and is basically a burden on the society.

Nuclear is not a panacea but railing against the technology because humans goofed up, then refused to pay more attention and do things correctly in the future is an irrational stance.

Same reason why I am absolutely terrified of playing any instrumental music on Spotify. I let one track run till the end and my next 2-3 discover weeklies will be filled only by instrumental music.

Buy commodity computer parts and build a PC. If it's strictly storage, then a cheap pentium/i3/Ryzen 3 would suffice. If you want to do more, then spec for a higher CPU. PCPartPicker should be able to give you the list of compatible parts.

Then depending on what you want out of the device, do one of the following -

1. If it's strictly storage and you want the niceties of ZFS, install TrueNAS 2. If you want expandable storage without ZFS + VMs + Docker with a nice GUI and are willing to pay for a license, install Unraid. 3. If you want ZFS-backed storage + VMs and no docker, and maybe some datacenter-style capabilities, install Proxmox 4. If you want all of the above and are willing to manually configure everything, then install Ubuntu Server + Cockpit for managing the machine headless + Portainer for managing a large number of docker containers.

I don't think anyone should have to resort to building multiple NAS and distributing them geographically. It's 2022 and this an extremely esoteric thing to do.

Given how giant companies (especially Google) behave, I do not feel it is an esoteric thing to do. Anyone with even a modicum of technical skill can roll their DIY NAS. For the non-tech savvy, buying off the shelf is fine.

Should not be needed in real life but this is the world we live in where robots have way too powerful ban hammers and the only way to get any support is to make sound in public forums or have a large social following.

That sounds very interesting. Have you documented it somewhere?

Twitter is not the only place for discussion. And people do not vote exclusively based on public discussion. Indians have tight knit communities and how the community votes is a much major influence on the elections than what is being discussed on twitter/facebook or even national TV.

Case in point, Indians might elect BJP for the central government but consistently have been booting them out at the state level for the past 3-4 years. BJP has actually lost power in multiple states since 2014.

Indian politics is way more nuanced than any discussion on an internet forum would ever allow for.

Why would an order go through with an expired credit card? Is this something like cash-on-delivery or card-on-delivery?