Another route would be AWS SES. They have a free tier. For managing the subscribe/unsubscribe process you could still use MailChimp or a self-hosted solution like Keila https://www.keila.io/, although I have no experience with this.
879 karma · joined October 28, 2019
Another route would be AWS SES. They have a free tier. For managing the subscribe/unsubscribe process you could still use MailChimp or a self-hosted solution like Keila https://www.keila.io/, although I have no experience with this.
I noticed they also offer an RSS to Email option, but I couldn’t find if it’s available on the Free plan. https://mailchimp.com/features/rss-to-email/
As a teenager I’ve been "traumatized" by Pascal, but I was lucky to find a refuge in HTML and CSS. After I got a job as a web designer I discovered Python and I realized that I could actually program.
The people who came up with this new syntax have different needs that I do and probably have a different kind of brain than mine.
I remember this from the book Alchemy by Rory Sutherland.
I was getting 2.2 tokens/s with the llama-2-13b-chat.Q4_K_M.gguf and 3.3 tokens/s with llama-2-13b-chat.Q3_K_S.gguf. With Mistral and Zephyr, the Q4_K_M versions, I was getting 4.4 tokens/s.
A few days ago I bought another stick of 16GB RAM ($30) and for some reason that escapes me, the inference speed doubled. So now I'm getting 6.5 tokens/s with llama-2-13b-chat.Q3_K_S.gguf, which for my needs gives the same results as Q4_K_M, and 9.1 tokens/s with Mistral and Zephyr. Personally, I can barely keep up with reading at 9 tokens/s (if I also have to process the text and check for errors).
If I wasn't considering getting an Nvidia 4060 Ti for Stable Diffusion, I would seriously be considering a used RX 580 8GB ($75) and run Llama Q4_K_M entirely on the GPU or offload some layers when using a 30B model.
Does the Python code you want to run use PyTorch? Is it related to LLMs or Stable Diffusion? Because those should also run on CPU alone. Not very fast, but good enough to play around with.
If you're thinking of renting a machine from AWS, I'm currently using a "g4dn.xlarge" instance which does support CUDA and can run Stable Diffusion comfortably. It's a bit pricey though and I had to wait and get approved before I could rent it.
I chose KeePassXC on the desktop. It has a database which you can move around, backup it. There is an app for Linux and MacOS 10.13+ (because of Qt5). For iOS, a quick search shows up something called KeePassium.
These apps seem to work with a certain database format (kdbx), so for sync you might need to place the database somewhere online (iCloud, as an example). Otherwise, you might not need syncing if you don't add 2FA accounts to your database that often.
At the end of each year, I run a Python script which through AppleScript creates an event at the beginning of each month that reminds me to send some papers to my accountant. Instead of just setting the event on the 1st of each month, the script chooses the first working day of the month, taking into account weekends and bank holidays.
Considering you can create multiple calendars, the script (with a cron job) could read the events in the "Exceptional" calendar and update the "Work" calendar.
Something similar might be possible with Google Calendar through the API.
For blocking ads, I run an Alpine Linux VM with dnsmasq and a massive block list. This VM masquerades as a DNS server.
Last fall I had to install a Letsencrypt certificate to have more websites working.
Lately I've been experimenting with Squid (caching proxy) to enable SSL bumping to get Wikipedia to work (they use TLS 1.3 and Safari 7 doesn't understand it). The experiment was successful, but I don't like how Squid behaves in other respects.
So it is possible to squeeze some more life out of these devices, but they remain limited because of Apple tying the browser to the operating system.
Hard to say with precision how much money you’d make if you put ads on your website, but this one (https://online-solitaire.com/), mentioned a couple of times here by its creator, is making $10K/mo. From the traffic you have I would guess you should be making at least 1K/mo and looking at the multiples on Empire Flippers https://empireflippers.com/marketplace/ then the selling price should be at least $50K.
So it’s definitely worth much more than $500 and less than $2M.
My mistake was basing my statement on past experiences using an older Firefox ESR version which would give "Content Security Policy: The page’s settings blocked the loading of a resource at inline (“script-src”)." when clicking on a bookmarklet for sites like Twitter or Github.
document.querySelectorAll('p').forEach(elem => {elem.style.color = '#000'})
You can turn this into a bookmarklet or run it from the console(necessary on sites like github.com which do not allow code running from bookmarklets).
Of course, it will not fix all your text color problems. To cover more elements like lists and headings, you'll need to add them to `querySelectorAll`.
document.querySelectorAll('p, ul, ol, h1, h2, h3, h4').forEach(elem => {elem.style.color = '#333'})
`#333` is what Firefox Reader Mode uses because it's less heavy than full black (#000).
First try to get your finances in order. If you still want to be an entrepreneur, being a freelancer or having a small limited liability company will give you the necessary time and administrative flexibility. Contact people you know (or worked with in the past) and let them know you are available for new projects beginning Monday or April 1st. In my case, former colleagues were the ones who got me work when my personal projects were still not making any money.
When your financial situation is more robust, try to start small and continue with the things that work. One of my biggest mistakes was to think that I was running out of time and in order to compensate for the lost time and income, only projects with big potential (high revenue) were worth pursuing. But projects like these are hard and might require capital, people and luck. Personally, I've had better fortune picking projects which were small, were in a niche where I had some expertise, and were proven to make money(I'm not very interested in projects where I don't know with some certainty how much income they produce).
I'll have to see if it can be "designed" with userChrome.css or something and I'll give it a try.
There's even a Sublime Text plugin to generate this text.
And a bonus anecdote:
"At a party given by a billionaire on Shelter Island, Kurt Vonnegut informs his pal, Joseph Heller, that their host, a hedge fund manager, had made more money in a single day than Heller had earned from his wildly popular novel Catch-22 over its whole history. Heller responds,“Yes, but I have something he will never have . . . enough."
I spent one summer in high school just to learn how to upload files via FTP. I would go once a week to an internet cafe (no internet at home) and I'd try to use shareware software(FileZilla did not exist yet) or upload files with Internet Explorer to some free hosting service, but nothing would work. When school started, a CS teacher overheard me asking a colleague about uploading files and recommended I use Total Commander's built-in FTP feature. From that day on, I knew how to upload files.
I honestly don't know what to recommend. I'd say use Linux because a lot of tutorials are doing that. But Linux might bring even more troubles, or you could be exchanging one set of problems for another. macOS might be a good compromise.
For me it was difficult learning how to use computers and how to program them. But I find it rewarding to solve problems and that gets me through the pain.
Also, when it launched, GitHub didn't really have a design to write home about, at least, in my opinion and according to these screenshots: https://github.blog/2008-11-03-github-code-search/
My experience was something like this: I had a small niche website while having a job. This website was making less than $50/mo from ads. When I quit my job to start an online business, I wanted to make something new, a community. But I had no experience with that type of website. Not only did I spend a lot of time (12 months) building the "product", but I also had no idea how to market it and no idea how to make money from it. Eventually, I started making small websites again, similar to the one I already had. I used the same platform, so a new "product" was ready in less than 3 months, sometimes less than a month. Surprisingly, at least to me, these small websites grew bigger and bigger, so they have become my source of income.
After all of this, the lesson for me was to stick to what works and what I can do.
Regarding point 6, I also have that problem, so I try to see if I can offer something more, different or if I can remix it. If not, I move on. I don't find it satisfactory to just clone an existing product.
I would have to click a few times before it would register one click. Very annoying. Just like in this post, the problem was the switches. But in my case, the switches were micro switches. Not knowing where to order the correct parts and having no soldering skills, I decided to pry open the micro switch. Luckily it was possible.
Taking the metal cap off, inside the switch there was a rubber membrane and a flexible metal disk. The disk was similar to a jar lid. The edges would touch the circuit board, and pushing the center of the metal disk, it would make it touch some conductive part of the board and it would "close the circuit".
The problem was that the flexible metal disk inside the micro switch was dirty. Oxidized, probably. I cleaned it, put the switch back together and the mouse works perfectly.
Joking aside, if enough work is put into it, SVG might be a solution one day to build a website. But right now it causes more problems than it solves.