ZeroNet Blogs

Static ZeroNet blogs mirror

Kaffie's Blog

I apparently just write about my search engine. But I do more than that, honest!

Twilight Princess HD

- Posted in Kaffie's Blog by with comments

Just picked up the new Legend of Zelda: Twilight Princess HD for Wii U. Haven't actually played it yet, and I'm not exactly in a rush since I've already played the original. In fact, this is the third version of the game I own. And I bought them all for different reasons, haha.

The original gamecube game I bought and played since it was new. I'm a huge Zelda fan, so it only makes sense. TP was pretty hyped up as well. I bought the Wii one (which actually came out earlier than the gamecube one, despite me picking it up later) so I could hack my Wii. IIRC it was the only way to do it at the time. I never bothered to beat that one, since it's essentially the same game, but has the content flipped, and added gimmicky motion controls. But nonetheless it's sitting on my shelf.

And now this new one. I wasn't actually planning on getting it. But it comes with an amiibo figure which will apparently be needed for the new Zelda Wii U game coming out later this year. And I was planning on replaying Twilight Princess soon anyway. The figure is pretty cool, no complaints there. And from what I've seen in screenshots and trailers, the game looks nice. Perhaps they've given it the MM treatment, where they make various improvements. I imagine they probably did something similar to how Wind Waker HD was, but I haven't had the chance to play that one yet.

I'm unsure of whether the game will work on my Wii U, which is partly why I'm hesitating to play it. See, I'm on FW 5.5.0, which isn't quite the latest. I accidentally updated it like a dumbass, and then all the cool hacking stuff came out for the one I had just updated from. And I feel like if I update it again, the same thing will happen! I feel like I'm gonna have to update the Wii U anyway when my brother and his girlfriend come to visit, since they'll probably want to play some other games. So perhaps I should just bite the bullet and update. After all, the IOSU exploit is possible on 5.5.1 as well. So maybe it won't be a big deal?

In Kaffiene related news, I've added a few more sites (like to the new ZeroWiki), and I've added a donate page, that lists my namecoin and bitcoin addresses. Namecoin so that I can keep the kaffie.bit domain, and perhaps register kaffiene.bit. Bitcoin, so that I can maybe bring in some money for doing this, since I'm currently unemployed :(.

My medical expenses aren't going to pay themselves... But on the bright side, I've managed to pull myself together enough to schedule an appointment with a doctor who might actually be decent, rather than just an idiot who wants to take my money for doing nothing.

Either way, this has been a bit of a deviation from the typical Kaffiene updates, since this is my personal blog after all. So hopefully that won't scare all you guys away. And I did promise gaming related stuff in my first post. So there's a bit for you. Perhaps I'll talk about A9LH and 3DS hacking in my next post.

As always, thanks for reading <3

Install Kaffiene?!

- Posted in Kaffie's Blog by with comments

That's right. I've added an experimental feature to allow visitors to install Kaffiene to their browser. From my tests, only Firefox worked. So it, along with any derivative browsers should work fine. Safari doesn't support the feature unfortunately. ~~And I couldn't get it working on Chrome, despite chrome saying it supports the feature.~~ So if Safari and Chrome users care to chime in and help figure out what the issue is, that'd be great.

Edit: I've managed to get it installing for both Opera and Chrome. The auto-detect and auto-install scripts have problems within the sandboxed Zeronet iFrame, so I've removed them for the time being. There's now an install page with browser-specific instructions located here.

I also began looking into how to make Kaffiene cloneable. I tried enabling the option, but upon signing it was removed... I'm trying to figure out how to just make it duplicate the whole site, index and all. So that anyone can easily click 'clone' and start modifying the index and site.

In unrelated news, it was my birthday yesterday, which is why Kaffiene's development has been at a crawl lately. Wasn't feeling too great though. So... happy birthday to me?

Edit: I've managed to make it cloneable. Editing the content.json directly didn't work. I had to edit it, and then hit 'save site settings' in the sidebar. Then sign and publish. Cloning the site will copy everything over. Just remove the Kaffiene specific parts and edit the index and such as you please.

Number Crunching

- Posted in Kaffie's Blog by with comments

In a very google-esque move (I guess I'm following a trend here) I went ahead and added a display for how fast the search was conducted. And you know how discover mode lagged a bit (along with blank searches)? Well I managed to optimize it a bit more. Now every search is running under 1 second for me. Whereas before, a blank search with discover mode would take over 5 seconds!

Along with that, last night I added an FAQ answering some common questions, as well as providing a bit of info I wasn't sure where to put.

I also added links back to the Kaffiene homepage in the various side pages. It's a colorful "Kaffiene" link in the top-left corner.

That's all for now.

We're finally different!

- Posted in Kaffie's Blog by with comments

Thanks to 'thatlurker', Kaffiene has finally got a new UI. I made a few tweaks to what I received, but I personally think it looks nice. More google-esque. I also went ahead and added a visual feedback for discover mode. The tags turn pink when you move the cursor over them, to indicate they're clickable.

I added a few sites, even though at this point I'm quickly being outpaced by ZeroSearch. Quality over quantity though, right? Hopefully I'll make a crawler soon, but we'll see.

I reworked the 'why' page and rewrote some stuff. Removed the large images.

I also redid the random search. It now randomly grabs an actual tag and uses that, meaning that on average more results should pop up and the terms shouldn't be as weird. This also clears up a lot of space on the site, since before the random tags were almost a duplicate of the index.

Sorry it isn't much, but disaster struck a couple days ago and my 2TB external hard drive had corrupted or something. I've been spending a lot of my time trying to fix it and recover the data. Which is why Kaffiene hasn't really had much development besides these small tweaks and changes.

I also want to give a shoutout to Censored Search, the newest ZeroNet search engine. It was cloned from Kaffiene and is exactly the thing I'm trying to encourage. Instead of ZeroNet sites, however, it searches clearnet sites. Not a direct competitor, but still something the community could see a lot of use from. You can find it here. I particularly liked how he kept the publicly accessible index. I do think sharing information and making it easy to access important aspects of sites is a key element of ZeroNet, and one I'd like to see expanded upon. The clearnet is typically the opposite; keeping everything secret. I really enjoy the openness here on ZeroNet.

I hope you all enjoy these latest changes. Thanks for reading!

A Kaffiene Update

- Posted in Kaffie's Blog by with comments

Finally got around to pushing out another Kaffiene update! Just a few quick ones this time. I moved both the index file and the checking script to the front page, so that they're easily accessible. I also went ahead and went through the early sites in the index that ZeroSearch added back in the day and retrofitted them with tags. However, much to my surprise, there were a lot of dead links. So I tagged them as 'old' and 'dead' as appropriate. Not all 'dead' tagged sites are inaccessible, but they're more or less useless. Whereas 'old' just means that I didn't see them updated for a long time, but if it doesn't have a 'dead' tag, it might still be a bit useful.

I'm wondering what I should do with these dead sites. Some are just a basic 'hello world' type of thing (why would someone add that to the search engine?) while others are just a personal blog with 2 posts about "I'm on zeronet!". And others still are a customized homepage with nothing really relevant besides a bit of flair/customization. Should I keep these sites? Remove them?

To see what I'm talking about, I'd like to introduce a new feature: Sharing searches! On every search, next to the results count there'll be a share link. Just 'copy link' or click it and copy the URL in the address bar. Do keep in mind that the url itself is related to how you arrived. Accessing Kaffiene through a proxy will net a proxy url, accessing it with the chrome extension will give a link accordingly.

Here's the link showing the old sites. As you can see, it automatically went with the standard local address and port, along with the domain name.

Alongside this new feature, I have fixed the search results for alternative zeronet users. Users with the chrome extension should now have proper links (rather than just pointing to the local address), proxy users should have links that go to their proxy, and so on. Please let me know if there's any issues.

Lastly, the site count issue. There was a lot of drama recently, but it's all mostly sorted out. However, Kaffiene did pass 300 sites indexed, which is why the 'low site count' message has moved to be linked from the word 'unique'. However, I noticed that in response to this, ZeroSearch has woken once again and added seemingly spam links to it's search. A quick search for 'ZeroBlog' will show that. To give credit, they do appear to be distinct and it seems ZeroSearch has more or less removed the majority of duplicates. However, those search results are entirely useless. Which 'ZeroBlog' should you click? There's no info, and some are actually useful while others are just 'hello world' type stuff! To me, this is the mark of failure in a search engine. A site that only cares about index count and completely disregards the quality of results. I've been hesitant to add brand new blogs, because I've seen first hand how frustrating it is to have dead links or empty blogs pop up in results.

I'm still trying to figure out a solution to this problem, but I certainly know that blindly adding 'ZeroBlog demo' as the title and just spamming a bunch of new blogs is not the answer.

And, in favor of enhancing search results, I've cleaned up the 'random terms' list. In the future I plan to have it automatically grab a tag from the index. But for now it's just a big list.

Thank you all for using Kaffiene and supporting quality over quantity. <3

New Kaffiene Features!

- Posted in Kaffie's Blog by with comments

Kaffiene got some improvements today. I added a 'discover' mode. To enable it, just check the checkbox next to the search bar. It allows you to click on tags and search for that tag. This makes it easy to find related content. It's disabled by default, since it can increase the wait time for larger searches.

I also added an option to search via regex. To do this, simply type "[r]" without quotes, then a space, and then your regex. When searching this way, terms separated by a space will be treated as separate regex searches. results will adhere to all of the regex queries (meaning it's "and" not "or"). The default search works in an opposite way: it'll pull up results that include any of the terms, rather than all of them. It's still possible to do an 'or' with regex, just don't have spaces and type something like: "termA|termB". This allows you to make powerful search queries. For example, if you want to search for termA or termB, but ensure that you also find termC, you'd search: "termA|termB termC". You can do other regex searches as well: "(forum|discussion|talk) (french|russian)". Any valid regex should work.

I also overhauled the searching itself, and hopefully cleaned up some of the redundancies that were left in when I cloned ZeroSearch. So that should make the searches a bit faster.

You can also now do reverse site lookups. Simply enter the address of the site. For example, this blog would be "1N6zp6jCXPBktNMPfe7UJBpQGyfCq7k2M8". Just enter that, and the site name/tags will pop up. It's a handy way of ensuring your site is in the index. But this will become even more useful in the future, when merger sites are added, and I create a query for bots/sites to get a plain-text or json version of the results. The possibilities are endless!

I removed the Roboto font that was taking up a whopping 30kb. It was only really used for the title, so I removed it. I'll probably replace the title with an image logo, so it looks nice.

Lastly, I fixed some of the links around the page. I noticed I didn't have them linking properly (as ZeroNet expects) and it was causing some weird errors. So that should be fixed now.

As usual, please continue to enjoy using Kaffiene. Thanks :)

On the topic of index counts

- Posted in Kaffie's Blog by with comments

It has recently come to my attention that several competitors to Kaffiene have been stealing recently indexed sites from the Kaffiene index, and not bothering to check the results or ensure quality. While I don't have any say over what these other sites do, I can call them out on their blatant dishonesty in an attempt to get users. In the spirit of competition and honesty, I've provided a write up over on the Kaffiene site about this issue. I've provided accurate site count numbers for both competing sites (ZeroSearch and Bwoi) as well as provided a script to check for duplicate entries and return the actual unique count. I also went ahead and made the Kaffiene site index public, changing the file extension to a '.txt' and provided a direct link to it on the site.

I hope that people can focus more on quality and accurate search results, rather than listen to dishonest reports on how many sites are indexed. And I take this as an opportunity to encourage both ZeroSearch and Bwoi to run the included python script on their own indexes and hopefully remove duplicate sites, improving the experience for users of the various search engines.

You can find the complete write up over on the Kaffiene search site, here.

Thanks for reading, and please continue to enjoy Kaffiene.

As you all may be aware, the only search engine for ZeroNet is ZeroSearch. And it's kind of crap (sorry to the dev!). It's capped at how many search results pull up, so you have to keep clicking search. It doesn't have descriptions or provide the tags for sites so you don't know where you're going. And, to be honest, it lacks a lot of newer sites.

Frustrated with these problems, I decided to clone the site and start working on my own. I removed the fancy 'request' system, since it didn't really do much, and I had problems with the ZeroNet functions. I've fixed the result count, added a display for tags, added all of the requested sites from ZeroSearch's queue. And I've added a few more sites that I was aware of. I also took the liberty of changing all of the site tags and names into terms that will pop up when you hit 'random', rather than just have the frustratingly short list of 5 or 6 terms.

I'm planning on having it automatically include sites that users add, and perhaps find some way to grab all the latest additions of 0list, 0name, etc automatically (though that might have to wait until merger sites arrive).

I've never worked on a search engine before, so bear with me while I figure it all out. Hopefully you all enjoy this latest project.

You can find it over at search.kaffie.bit, or if you want the direct link (in case I fail at namecoin and do something dumb) you can find that here: 1Mr5rX9TauvaGReB4RjCaE6D37FJQaY5Ba



- Posted in Kaffie's Blog by with comments

Kaffie.bit is now registered and points to this blog! Yay. I don't know how long I'll be able to support it, since I'm still new to this whole namecoin thing. But for the time being feel free to use the domain to get here. Big thanks to cannonfodder and glock who helped me set it up. If you'd like to help keep Kaffie.bit pointing here, feel free to send some namecoin my way: N4Bee3vzgpQnjC3Kah1BQENLZ6fatsbUkq