Herewith I finally start the documentation of one of my smaller projects. Actually, I wanted to have started this blog article several months ago, but a move to a new apartment got in the way. This is associated with a lot of work, which I unfortunately had to subtract from the documentation. Now I've finished the roughest renovations and only have a few small tasks left to do around the apartment. So far, I've finally unpacked most of the boxes, my cat has settled in so far, and I can get back to working on the Meklet project at my bedside. Unfortunately, I don't have a desk yet. I'm dividing this blog entry into three parts. The past, the present and the future.
The Meklet search started as a hobby project and was nothing more than to simplify my own life. I was overall dissatisfied with what the previous search engines offered and wanted something that would take me less time to search for information. Overall I had a few points that I didn't like. Too much spam, too many results tailored to the search engines and there was no suitable alternative for me. There were already some requirements that a search engine (which Meklet is not) could already see from me.
So I knew in very many cases where to look for what. So when I was researching a video, for example, I usually looked at the biggest provider first. The same was true for products, news and computer games. Nevertheless, I kept getting search results that didn't interest me at all. Worse, they even distracted me from the search. So if I searched for "monkeypox" because it was a hot topic, I would start with this new topic in Wikipedia and then eventually end up on medical websites. I didn't want to get a link from a small search engine optimized blog that showed me its personal opinion on the subject.
So the search engines are only intermediaries between me and the information I wanted to see. So I wrote something very simple in HTML and tested it locally on my MacBook for a very long time. I now directly accessed the search indexes of the websites I was interested in. Only when it got a little bigger, I uploaded it to a hoster. First to
Google Firebase, then to
Neocities. Since I didn't have much time, I worked on the project every now and then for a few hours a week, but never so much that I was only busy with it. I didn't want to invest all my free time in this one idea. Because it didn't seem important enough to me at that time.
My first name I chose for the project was "SearchBox", but I didn't like it pretty quickly. Overall, the name was too generic and boring. At some point I came up with
Meklet, which means "search" in Latvian. My grandfather came from Latvia to Germany after the war and started a family here. So the term also represented a reference to my own past, because my grandfather died very young and I never got to meet him. In addition, it was also a
character in the computer game Atelier Sophie: The Alchemist of the Mysterious Book, which represents an artificial intelligence and so seen also something to do with "a search". But for that I still have to play this game. In summary, I had a more interesting background with it than with Searchbox. As a framework I used Bootstrap, but I threw it out in this version.
There are some differences in this version that I will explain right away. As I noted in the previous paragraph, I used the
Bootstrap framework. In my opinion, there is nothing at all wrong with using a framework for an idea. An idea needs to be built quickly so that you have something to show after at least three days. It doesn't need to look good or work perfectly, it just needs to explain the core features of the project well. Unfortunately,
Elon Musk bought Twitter and with that, so did the Bootstrap project and its team. All alarm bells have been ringing for me in recent weeks, because I wouldn't put it past Elon to convert the Bootstrap framework into a commercial Bootstrap. Who wants to stop him? Anyway, it animated me to release the next version. Yes, I think JavaScript is quite good, but I don't want to rely on it working well on every computer. So my decision was to convert everything once to vanilla HTML.
I have been an Apple user for almost half a year. I had used Linux and Windows for many years, but always had problems with the whole vibe. I don't want to go into it here, because all three systems have their advantages and disadvantages. Everybody should decide that for himself. The new Meklet version is very much inspired by the search on the iPhone. Roughly. There is only one darkmode for now and everything is kept minimalistic and simple. I had emojies and the Bootstrap typical colorful design before. I didn't like it at all anymore. It was all so colorful and terribly childish. I don't want to copy the Apple design one to one, but I want it to fit into the system environment without being too intrusive or annoying. Therefore I kept everything in black, white and gray. But I'm not a great designer or CSS developer and this has to be improved a lot in the next versions. Maybe even by a person who can do it much better.
Since I now could not fall back on a complete CSS I had to develop something myself. I named my CSS "Framework" after my grandmother Irmgard. She comes, as you can see from the name, from Germany. The CSS and also the Meklet project are under the MIT license. This was also something I had not taken care of yet. But since I don't have the time at the moment, I won't promote the project on social networks or upload it to Github for example. If you want to create your own hoster with a meklet copy and add your own search strings please do it yourself. For me this is still a hobby, although I'm not averse to offering meklets as a commercial product. But these are all unlaid eggs and if it will come so far I can't say under any circumstances. It will take some time until something is in the bag.
One problem with vanilla HTML is that you don't have all the great features anymore, like being able to show and hide the individual categories with a little arrow. I decided to use anchors, which should limit this problem a little bit. Whether this works as I imagine it I can not say yet. But since users are almost never interested in "everything", they can now link to individual categories. This should also work as a bookmark in most browsers. Otherwise I continue to collect links and add them all by hand. I still have some interesting sources here that I want to edit. This is cleaner than farming with bots. Less is sometimes more.
There are some points I will work on in the next weeks. Since I want to connect Meklet more with the Apple system, I will work more intensively with
Shortcuts,
AppleScript and later maybe even
Swift. I have a lot of ideas, but I don't know if they will work, because I am a short term Apple user. And yes, I will put my focus on this system first for some reasons. At the beginning I had focused on a mobile website of the project but I will change that as well. I want to develop something good for desktop first that other people can work well with. The branding, logo design, CSS design and all the other little things will be more focused on in the next versions. I will have to buy a domain to make this toy project look at least half serious.
I have many ideas here, most of which I will try out. But they can also quickly fall under the table again if I just don't like them. So far nothing is set in stone here. If you want to get more information about the Meklet project you can
subscribe to the RSS,
download the shortcut or have a
look at the statistics.
Robert