? Editing: Post:21.body Save Delete Cancel
Content changed Sign & Publish new content

klu9's ZeroBlog

Just a user, not a coder.
Interested in ZeroNet's potential for a web less vulnerable to:
° regulatory capture (e.g. copyright laws dictated by lobby groups)
° takedowns (e.g. DMCA, EUCD, Article 13, domain seizures)
° blocks (e.g. ISPs, GFW)
° net partiality (e.g. preferential treatment for privileged content providers)
° corporate acquisitions, changes, neglect or demise (e.g. GeoCities, del.icio.us, Flickr, Tumblr)
° individual change, neglect or demise (e.g. linkrot; webmaster losing interest, disappearing or dying)
° money-making shennanigans supposedly justified by the cost of providing a service (e.g. advertising, tracking, privacy violations)
° bloat (wait for a 10MB page to load just so you can read a 10KB article)
° connection problems (as 0net can cache sites)

klu9's media share
The Papers
ZeroMe

Follow in NewsfeedFollowing

TOC by date
TOC by tag

Latest comments:

WebP for comics with free tool CbxConverter

on May 27, 2019
tag: images comics newb 0net

A post on ZeroTalk about uploading comics to ZeroNet led to discussion of alternative formats, like DjVu (which I used to use to convert massive file-size JPG scans of textbooks into files that were a fraction of the size and used a lot less of my computer's RAM than PDF, and unlike JPG can contain an OCR/text layer).

I thought "Hey, maybe WebP would useful for comics." And it turns out that (obviously) I wasn't the first to think that. Many comic book viewing programs already support WebP.

I tried converting a comic book page from JPG to WebP myself, using an image editor. But then I found there's a handy tool to convert entire comics to WebP.

CbxConverter is a small basic GUI program for Windows that takes 'standard' comic book files like .cbz and .cbr (basically .zip and .rar files usually containing JPG scans), converts the image files inside from JPG to WebP, and produces a new .cbz with a significantly smaller file size. Just drag & drop the cbz/cbr onto the program and press "Start conversion". After a few minutes, you get a brand new .cbz file at a much smaller size, ready to read in your comic book viewer (in my case SumatraPDF).

CbxConverter.0.12.2

In my brief testing, here's what I got with it.

No. Type Details Size Sample page
1. Original cbr containing jpgs 60 MB Jpg: 1,630KB
2. Jpg Recompressed Q:75 only 1 page Jpg: 1,260KB
3. WebP cbz Default (Quality:75) 30 MB WebP: 821KB
4. WebP cbz Quality:50 21 MB WebP: 585KB
5. WebP cbz Resize:66%, Quality:50 12 MB WebP: 322KB

(Re 2, I extracted a jpg from original and manually recompressed it as jpg in my image editor at quality 75 to see how that would compare to CbxConverter WebP at 75. I only did that for one page.)

I literally could not tell the difference between No. 1 (60 MB), No. 2, No. 3 (30 MB) and No. 4 (21 MB), at least viewing on my screen and at "Fit Width to Screen". Even at full size most of the time I couldn't tell the difference.

CbxConverter 0.12.2 doesn't allow a free choice for resizing, just a selection of 66%, 50%, 33% or 25%. But with this particular comic, 66% is only a little under my screen's width, so it's a good size. In fact, in a few places it even looks a little better than the original (when the original is scaled to fit my screen's width). So for me, No. 5 (12 MB) is practically as good as No. 1 (60 MB). But less than one quarter the file size :)

(For coders, the source is available at the bottom of CbxConverter's homepage.)

Now if only I could find a nice handy tool like CbxConverter but to convert entire comics files to DjVu, so I could compare.

Read more

Getting a vanity address for a zite

on May 06, 2019 · 7 min read
tag: 0net newb URL blog

Standard ZeroNet website addresses, being based on Bitcoin addresses, generally look like a long string of random numbers and letters, beginning with 1. E.g this site's:

  • http://127.0.0.1:43110/1Ctjcvfovq6d1JWEGvM1JovCym4H6Es33L/

But you may have noticed that a few are not entirely random. E.g. ZeroHello's has 1HeLLo at the start:

  • http://127.0.0.1:43110/1HeLLo4uzjaLetFx6NH3PMwFP3qbRbTf3D/

You can create an address with your own choice of characters at the start by using a software tool called vanitygen. So here I'll write about what I did to get a vanity address for a new zite.

I based what I did mostly on this guide:

However, maybe because that guide is a few years old, or perhaps because I'm just not good enough at all this stuff as the original writer, I found it a little confusing or inconvenient at certain points, and I ended up doing a couple of steps differently. (I thought I might find the issues clarified in the comments, but even though the page says there are 6 comments, I can't see any.)

Part 1: Creating a vanity address

Note: I did this on Windows. There will be slight differences on other operating systems.

1. Get vanitygen for your operating system.

Available through the links on the Vanitygen homepage. I'm on Windows so I used the vanitygen contained in this zip file.

2. Open a terminal emulator / command (CMD) window, and direct it to the location of vanitygen.

After unzipping the file, I opened a command window. ("Start" > search cmd.) Then in the file manager, I copied the path to the vanitygen folder:

  • C:\Users\klu9\Downloads\vanitygen

and then in the Command window right-clicked and selected "Paste". (Ctrl+V doesn't work in the command window.)

3. Enter the vanitygen program name and the pattern you want your address to begin with, starting with the obligatory 1.

All Bitcoin addresses have to begin with 1.

So in my case I entered into the command window:

  • vanitygen64.exe 1Papers

01vanitygen.png

  • Vanitygen, a few hours after it started working.

Don't make the pattern too long, as each additional character increases the time needed exponentially (e.g. if I had chosen 1ThePapers, those three extra letters could have made the process last for days). Vanitygen requires a lot of processing power from your CPU, so before I started the process, I closed every other program.

4. Wait while Vanitygen creates your address.

At the start, Vanitygen told me it would take 10 hours to get 50% of the way to creating the address. In the end that was about how long it took to complete the process fully. During that time, the task manager showed my CPU being used at 99% the whole time. Obviously it's not ideal to do other tasks on the computer while vanitygen is doing its work, so plan this for a time when you won't need your computer for other things.

02vanitygen.taskmanager.pix.png

  • Task manager showing CPU at 99% while vanitygen is working. Full size

5. Save the result.

At the end of the process, the command window showed me:

  • the pattern I had entered ( 1Papers )
  • the complete address it generated based on that pattern ( 1Papers8S7fe6LEsa7BwVVywWVbMpUjgwx )
  • The private key ("privkey"). This is what you will need to "prove" you own this address.

Keep your private key safe and secret. This is the only time you will receive it, and if you lose it, it's lost forever.

This part (creating the vanity address) was, although time-consuming, very simple and according to the guide I linked to earlier.

03vanitygen.result.png

  • Vanitygen showing the results

Part 2: Using the address

I wanted to make a fresh blog based on ZeroBlog but with the 1Papers address. Here's how I ended up doing it, which may or may not be the best way to do it.

1. Create a folder with your new address.

So in my case inside the folder

  • C:\Users\klu9\Downloads\ZeroNet-win-dist\data\

    I created a new folder with the address name, resulting in:

  • C:\Users\klu9\Downloads\ZeroNet-win-dist\data\1Papers8S7fe6LEsa7BwVVywWVbMpUjgwx

A little confusion on my part

Here's where that guide gets a little confusing for me. It says to go into the folder of an existing site and "copy everything". Really? Everything? I don't think I want everything. Wouldn't that make another version of the existing site with all its associated content and data? Whereas I want to make a fresh, empty site. If I copy everything, am I going to have to manually delete lots of stuff to make it a fresh, empty site? If so, which stuff? No answer there.

E.g. I took a look at the folder for ZeroBlog (\ZeroNet-win-dist\data\1BLogC9LN4oPDcruNz3qo1ysa133E9AGg8) and just its \data\users folder contains 2,606 files in 1,303 folders.

So "everything" would presumably include:

  • all the posts made
  • all media (images, videos etc)
  • all data on users who've visited the zite
  • other?

So, should I instead clone ZeroBlog from scratch and do something with the result? Would doing that create a new key that is then deprived from the pool of available keys?

2. Clone ZeroBlog.

Rather than copy everything from an existing site with data already in it, I decided to make a fresh clone of ZeroBlog.

  1. Go to the ZeroHello left menu
  2. Find ZeroBlog. (If you find lots of similarly named sites in your list, the one I chose was the one by ZeroNet creator @nofish:, whose URL is http://127.0.0.1:43110/Blog.ZeroNetwork.bit/ when you hover over it in the ZeroHello left menu)
  3. Click on the three-dot menu to right of site name and select "Clone"

04.0zerohelloclonezeroblog.png

That created a fresh empty ZeroBlog with the following address:

  • 16xTLL1n8pNfdEx8888CivfHGwFj926h8r

Which opened automatically in my browser, showing the "Congratulations" message.

04zeroblog.clone.16xTLL.png

  • New clone of ZeroBlog, with 16xTLL1n8pNfdEx8888CivfHGwFj926h8r as its address. Full size

3. Copy everything from new blog folder into the folder of your vanity address.

So in my case I copied everything from

  • C:\Users\klu9\Downloads\ZeroNet-win-dist\data\16xTLL1n8pNfdEx8888CivfHGwFj926h8r

into

  • C:\Users\klu9\Downloads\ZeroNet-win-dist\data\1Papers8S7fe6LEsa7BwVVywWVbMpUjgwx.

4. Open the new address in your browser.

I opened http://127.0.0.1:43110/1Papers8S7fe6LEsa7BwVVywWVbMpUjgwx/ and got the "Congratulations" message. Yay :)

05zeroblog.clone.1papers.png

  • New clone of ZeroBlog, with 1Papers8S7fe6LEsa7BwVVywWVbMpUjgwx as its address. Full size

5. Open the new site's sidebar.

Click and drag the ZeroNet icon (near the top right of the page) to the left, and let it go.

06zeroblog.clone.1papers.sidebar.1.png

6. Scroll down the sidebar and click "This is my site."

07zeroblog.clone.1papers.sidebar.2.png

More stuff will now appear below in the sidebar.

7. Edit your site's details in the site sidebar.

In my case:

  • SITE TITLE > The Papers
  • SITE DESCRIPTION > UK newspapers blog mirror

Then I clicked "Save site settings".

08zeroblog.clone.1papers.sidebar.3.png

  • Sidebar, additional features after clicking "This is my site". Full size

8. ???

Here the guide gives an instruction that was unclear for me.

Open content.json with a text editor and check that no domain it set.

If it is, delete it.

I don't know what/where such a domain is, but my site works, so I guess I'm OK.

9. Add the private key.

The guide's next step says to shut down ZeroNet and then run a terminal / command window to sign the site. But the sidebar has a button for this. (Maybe it didn't back when the guide was written three years ago.) The sidebar also has a link "ADD SAVED PRIVATE KEY".

(What about the identity address? 15i1fCD9BLVurrTuaN2boWoMX2urQd3CBe. That seems different from my usual.)

I clicked the add private key button and entered my private key.

Private key saved for site signing.

The button changes to a message "Private key saved." And also now there is a link "FORGOT". Oh, so maybe there is a way to recover your private key if you forget or lose it?

09zeroblog.clone.1papers.sidebar.4.key.png

10. Sign and Publish your new site.

I clicked "Sign and Publish". At first no reaction, but then some messages appear.

Content publish failed.

I've come to expect when creating a brand new zite. Basically it's because of the next message...

No peers found, but your content is ready to access.

A brand new zite has no peers yet, hence the "Content publish failed" message. Once a few peers visit the site, that shouldn't be a problem any more.

10zeroblog.clone.1papers.sidebar.5.png

11. Reload your new site and start using it!

At first I couldn't edit my new blog, but upon reload I could, and did a few basic things:

  • Changed blog title and description.
  • Made an "About this blog" post.
  • Added link to the About post in the blog description.
  • Already automatically "Following" but only posts (which I already know about; I'm the one making the posts). So I also selected "Comments" to make sure I'm notified of those in my ZeroHello feed.

11thepapersabout.png

  • Final result: about page with new blog with new address, following comments. Full size

After doing the above steps and adding some content, I publicized the new site a few ways to get some peers for it.

Conclusion

Well, that's how I ended up creating and using a vanity address, and it seems to be working. Let me know if I made any mistakes.

Read more

A rant ;) about intellectual property

on Apr 26, 2019 · 2 min read ·
6 comments

tag: not tagged

Today, the 26th of April, is World Intellectual Property Day. So I made a bit of a jokey post about it on ZeroMe.

That led to some comments there, including a bit of a rant ;) from me that I'd like to post here (with a few small tweaks & corrections), as it sort of sums up my views on the issue.

Read more

ZeroHello search & ZeroMe issues

on Apr 21, 2019
tag: 0net newb ZeroMe search

ZeroHello, the "start page" of ZeroNet, has a search box near the top, to look for content on:

  • any ZeroNet site you're connected to
  • a specific site, by adding the search query site:sitename

But I've noticed that when I do a general search there, it never returns results from ZeroMe. At least not from "official" ZeroMe, i.e. Me.ZeroNetwork.bit. But it does find ZeroMe posts through the alternative ZeroMe site/interface, Peeper.

I mentioned this in a comment on ZeroMe:

And kxobot pointed out that ZeroHello search can find results from Me.ZeroNetwork.bit, but only if you search that site specifically.

For example, if I do a ZeroHello search for just "favicon", I will get results from Peeper, ZeroTalk, people's blogs etc. From all over ZeroNet except for Me.ZeroNetwork.bit.

However if I do a search for "favicon site:ZeroMe", I do get results from Me.ZeroNetwork.bit. The same results that it found on Peeper.

So that's odd. But it may seem irrelevant whether the results come from Peeper or Me.ZeroNetwork.bit. But often in the past, when I clicked on a result from Peeper, a Peeper page would open but the content (post, comments etc) would not appear. Now that I'm experimenting further, I realize the content does appear on Peeper pages, it just takes longer to show up than it does when I load a Me.ZeroNetwork.bit page.

Hopefully one day ZeroHello search will include results from Me.ZeroNetwork.bit among the rest.

Read more

Remember to follow your blog's comments

on Apr 06, 2019 ·
1 comment

tag: 0net newb blog

I just remembered, thanks to only now noticing some week-old comments on my media share blog, that these ZeroBlogs (regular and MyZB both I think) will automatically follow posts on your blog (so you will see notifications of those in your ZeroHello feed) but not comments, so you will not receive notifications of those.

You have to remember to manually follow comments in order to receive notifications. In your blog's left-hand menu, click on the button that says "Follow" or "Following" and tick "Comments".

ZeroBlog.follow.comments.8bit.png

I also ticked "Username mentions" as well just in case, even though I don't really understand if it's necessary if "Comments" is ticked.

Read more

Experimenting with images: WebP

on Mar 29, 2019 ·
5 comments

tag: newb, images

When experimenting with replacements for animated GIFs ( part 1 & part 2 ), I (re)discovered the WebP format.

I looked into the animated version (like GIF), but it can also be used for still images, like JPG. There are claims it can make an image the same quality as JPG but with a file size up to 30% smaller.

I wanted to post a link to an image in a ZeroMe thread.

  • I did a DuckDuckGo image search
  • found the image I wanted (an illustrated book cover for Stig of the Dump ) and saved it to my hard drive
  • uploaded it to ZeroUp Unlimited

and then promptly realized "WTF?!?! 647KB! For that little JPG!?"

Here it is:
Stig - original JPG
stig.revealed.jpg - 647KB

I opened the JPG in Paint.net to recompress it and hopefully reduce the file size a lot. I clicked "Save as..." and then adjusted the quality settings.

Normally even just the default quality setting of 90 is enough to make a significant reduction in file size. Sometimes I take it down to 75. Only rarely, perhaps with a really large image-size, high-quality photograph do I take the quality setting down further, to maybe 60 or 50.

Here's what happened with this JPG:

Qlty File size Note
90. 632KB... "Well that didn't do much!"
75. 608KB... "Nor that"
50. 589KB... At this point, I can hardly believe how little the file size is reducing
25. 574KB... "Seriously???"
0. 553KB... I have never taken the setting down all the way to zero on any file... and yet this damn (small) JPG is still over half a megabyte!

But just the other day I installed the WebP plugin for Paint.net, so I decided to give that a go.

Here's what happened:

Qlty File size Note
90 84KB "Damn! I didn't even need to do any special tweaks."


Paint.net - saving image as as WebP

Paint.net - saving image as as WebP

I looked and looked and at normal viewing size, I can barely tell the difference between the 647KB JPG and the 84KB WebP. The WebP is only 13% the file size of the JPG. (Changing through the Presets like 'photo', 'drawing' etc barely made any difference in file size.)

Stig - original JPG Stig- WebP
Left: stig.revealed.jpg - 647KB | Right: stig.revealed.webp - 84KB

Conclusion

Obviously this kind of saving wouldn't happen on every type of image, but it still seems we really should be using WebP instead of JPG, especially here on ZeroNet where we're trying to save bandwidth and storage space.

One issue with WebP compared to JPG. By default, in my browser (Vivaldi), if I click on the URL for a JPG, the image opens in the browser. But If I click on the URL for a WebP, the browser downloads it and then opens it in the default local viewer for that file type. EDIT: Apparently this is an issue specific to how ZeroNet on Windows treats WebP, and an issue has been filed on Github. See this post on ZeroTalk.

PS this is also my first experiment with markdown tables. Nice to have the feature but I wish the spacing & alignment were better. And side-by-side images too! Although I couldn't get the captions to align with the images using markdown table syntax.

PPS I forgot that hotlinking images can be an issue on ZeroNet. Just because I can see the images hosted on ZeroUp Unlimited doesn't mean other people can. So I put a link to the source under each image. I've got to remember to be careful about that.

Read more

.mhtml format for saving webpages

on Mar 26, 2019 ·
1 comment

tag: newb web file upload

.mhtml is a format that saves an entire webpage's content (including images etc) in just one file. It's been around for years. I remember saving webpages in .mhtml back when I was on dial-up. It's also basically how e-mail messages can have fancy formatting, images etc.

Just in case you got a security warning when getting any .mtml files I've shared, let me say it: I'm not trying to harm your computer. .Mhtml is just the easiest way I know to save and share webpages. If you're unsure, please scan the file with an anti-virus.

I've been using .mhtml to save and repost the BBC's "The Papers" blog on my other blog klu9's media share, which has large images of UK newspapers' front pages, but only for one day. After that the BBC website deletes the images, for copyright issues I believe. You can compare the mhtml file I saved with the clearnet page from the BBC, from which all the images have already been removed.

For me, I save the page as .mhtml in the Vivaldi browser. And when I open the file in the file manager, it opens in Vivaldi straight away. (Oddly, it also appears as a download in Vivaldi's download panel.)

But when I open the .mhtml from within the browser (e.g. from posting it on ZeroNet), Vivaldi gives me a security warning about this file type and will only open it if I click "Keep".

(Why is a webpage saved as .mhtml considered inherently dangerous? Why is it any more dangerous than a webpage served as .html calling in God-knows-what resources? Someone better qualified than me can maybe tell us why.)

.mhtml been around for years, and browser support and treatment of the format has varied over the years. It seems it's not as easy and widely supported as it used to be. But I, as just a dumb user, have found no quicker & easier way to save and repost an entire webpage. I just click "Save Page as... MHTML" in my browser and then upload it to ZeroUp Unlimited and that's it. I don't have to fiddle around creating folders in my zite, deal with HTML coding or wonder if key content (like images) will break/disappear.

But as it seems there are issues with .mhtml nowadays, here I'll test it out a little in different browsers (on a Windows desktop):

Chrome


mhtml.chrome

  • Chrome does not open the link but treats it as a download.
  • It gives a warning message, saying this kind of file can harm your computer.
  • I can see no way within Chrome to open the .mhtml file in Chrome itself (it only offers to open it in my OS's default program for .mhtml, Vivaldi).
  • But if I right-click on the .mhtml file in the file manager, I can choose Chrome and it opens just fine.

Summary
Chrome can open .mhtml, but it will download it first, give you a security warning and, once you select "Keep", then you can open it in Chrome.

Firefox (Quantum)


mhtml.firefox

  • Firstly a brief ZeroNet error message pops up.

    Connection with UiServer Websocket was lost. Reconnecting...

  • But after a couple of seconds it says

    Connection with UiServer Websocket recovered.

  • Firefox does not open the link but treats it as a download.

  • It gives a dialog box to choose
    • whether to open or save the file
    • if to open it, what program to open with... and Firefox is not in the list of options. If I choose "Browse" and then find firefox.exe, it pause for a while and then sends me back through the whole "what do you want to do with this file" dialog box.
  • If I save the file, like Chrome, it will open in my default program, Vivaldi.
    • But if I right-click on the .mhtml file in the file manager, Firefox is not in the list of available options.

Summary
Firefox Quantum (i.e. recent versions) treats .mhtml as a file to download, and refuses to open it all.

K-Meleon

mhtml.kmeleon.1

mhtml.kmeleon.2

K-Meleon is based on an older version of Firefox.

  • Just opening my 0net blog in K-Meleon gives those "websocket connection lost / recovered" messages.
  • those "websocket connection lost / recovered" messages happen again when I click the .mhtml link.
  • Just like Firefox, it gives a dialog box: "What should K-Meleon do with this file?" With one difference: K-Meleon is among the options to open the file with.
  • But if I try opening it with K-Meleon that way, it opens a blank tab and that dialog box comes up again. And if I again choose open with K-Meleon, a new blank tab and new dialog box open again. And that can go on and on.
  • The same happens if I try going through the file manager, right-clicking on the file and choosing Open wth... K-Meleon. Blank tab and a dialog box again.

Summary
K-Meleon treats .mhtml as a file to download, and makes you think it will open it but doesn't.

Safari

As if!!!

Internet Explorer

Is that still on my computer? What do you know, it is!

mhtml.ie

  • IE just opens the. mhtml as it would any other webpage. No dialog boxes, no security messages, no treating it as a download. It just plains opens it the way it's supposed to.

Summary
Internet Explorer just works! Who'd have believed it?!

Conclusion

So it looks like .mhtml is not ideal for serving a saved webpage to other people: their browser may not support it, and they may get spooked by security warnings. I just wish I knew of a way as easy as .mhtml for saving and reposting. (Another advantage: I just share it via an upload site like ZeroUp Unlimited, so it doesn't take away from my blog's limited space and I don't have to learn about things like optional files just yet.)

Any (preferably easy) suggestions would appreciated. Perhaps even one that combines with the other thing I'd like to do, which is automate the process of saving The Papers blog article (on day of release, with images at maximum quality) and republishing it on ZeroNet.

What happens when you open an .mhtml file?

Let me know what happens for you, and what browser & operating system you're using.

Read more

Links with brackets in markdown and ZeroBlogs

on Mar 24, 2019
tag: 0net newb blog markdown links

Sometimes you want to link to something that has brackets (...) in the URL or the file name.

E.g.

  • https://en.wikipedia.org/wiki/Matrix_(protocol)
  • http://127.0.0.1:43110/1CiXRY9ATZSoZqBzwMfXEMsKtPRt2aQoF2/data/users/131Seivv6aH8vnohPmYPcPwPfZ1a5ubMMs/UK.Newspapers.Front.Pages-2019.03.24.Sunday-BBC.News.(for.desktop).mhtml

But this can be problematic on ZeroNet: in several places (this blog here, ZeroMe) when I tried to post the link in MarkDown (the easy code for website formatting, used here on ZeroNet and elsewhere), the result was a broken link that displayed the closing bracket rather than making it part of the link.

E.g. If I wrote in markdown:

  • [Matrix](https://en.wikipedia.org/wiki/Matrix_(protocol))

Then the page would show:

  • Matrix)

And the link would be:

  • https://en.wikipedia.org/wiki/Matrix_(protocol
    without the closing bracket. So clicking it would result in an error page.

Fortunately @glightstar: found the solution:

This is when you use % plus some numbers to represent certain characters to make sure the browser realizes it's part of the URL.

So to make sure links with brackets work, replace the URL's closing bracket ...) (but not the markdown tag's closing bracket) with %29. E.g.:

  • [Matrix](https://en.wikipedia.org/wiki/Matrix_(protocol%29)
  • [UK.Newspapers.Front.Pages-2019.03.24.Sunday-BBC.News.(for.desktop).mhtml](http://127.0.0.1:43110/1CiXRY9ATZSoZqBzwMfXEMsKtPRt2aQoF2/data/users/131Seivv6aH8vnohPmYPcPwPfZ1a5ubMMs/UK.Newspapers.Front.Pages-2019.03.24.Sunday-BBC.News.(for.desktop%29.mhtml)

And that should give you the properly displayed, properly working links below:

Note: this problem also happens when using the WYSIWYG editor in regular ZeroBlog to add a link with brackets, as I discovered when making this post on klu9's media share. In the little popup box, make sure to replace the closing bracket with %29.

For some reason, it doesn't seem necessary to replace the opening bracket with percent-encoding.

Read more

Tags on My ZeroBlog

on Mar 23, 2019 ·
7 comments

tag: newb 0net blog tags

I like the idea of tags for digital content like blogs (I know, I know: very 2000's of me). And, in contrast to regular ZeroBlog, My ZeroBlog (from which this blog is cloned) has tagging built in.

  • (See this post for a little more on the issue of the difference between ZeroBlog and My ZeroBlog.)

So for fellow newbs, here are a few points I've noticed about tags on My ZeroBlog.

Tags don't need commas to separate

Just use spaces to separate tags.
Out of habit from other blogging platforms, I put commas and spaces between tags, like so:

  • 0net, newb, blog

But My ZeroBlog was treating the comma as part of the tag. I think that meant if the tag was the last in the list and had no comma after it (like "0net"), it would be treated as a different tag from one with a comma (like "0net,"). So once I realized the issue, I had to go back and clean out all the commas. So now when I put in several tags, it should be:

  • 0net newb blog

Multi-word tags have to be done as one word

  • The above point about spaces & commas means multi-word tags (like "social media") have to be done as one word (like socialmedia). Otherwise, your post will get tagged with social and also with media.

Tags are case-sensitive

  • P2P and p2p are treated as different tags. So always keep the case consistent.
  • In the "TOC by tag" (i.e. list of all tags on your blog), tags starting with uppercase letters are all listed above tags starting with lowercase letters. (At first glance I thought the list was not alphabetized... until I noticed it is alphabetized, but twice: first tags starting with uppercase listed alphabetically, then tags starting with lowercase listed alphabetically.)

MyZeroBlog.tag.index.png

  • Full size
  • My tag index, after I'd cleaned up some things but before cleaning up others. Note the dual alphabetization (first uppercase, then lowercase) and case distinction (P2P and p2p are treated as different tags).

Sometimes tags appear replicated

Every now and then when I'm working on and viewing this blog, a post's tags seem to multiply and replicate. (At least for me, I don't know if that happens to visitors.)

MyZeroBlog.tag.overload.png

But don't worry: it's just a temporary effect that disappears when you refresh the page :)

Read more

Getting more space on ZeroTalk

on Mar 23, 2019 ·
2 comments

tag: 0net ZeroTalk newb space

Sites on ZeroNet that allow other people to do things (like make posts, comments, upvotes etc) have a limit on the space they allow users for that content and activity (to prevent spamming).

ZeroTalk, the principal forum on ZeroNet, by default allows new users just 100KB. You can see how much space you've used up just below the comment box, saying something like USED: 80.7K/100K.

When you get close to that limit, a message will appear there telling you to contact nofish (the admin of ZeroTalk and the creator of ZeroNet itself) to request an increase. His name links to his ZeroMail address.

ZeroTalk - running out of space.png
Full size image

But don't use ZeroMail: it won't work. (It didn't for me, at least.) Instead use ZeroMe. Write a request there mentioning him and make sure you format his name with @ and : before and after, so that the message will appear in his ZeroHello notifications feed.

E.g.

@nofish: Please raise my limit on Zerotalk. Thanks.

Using ZeroMail, I still had no response or increase in space limit after 12 days. But within 1 day of doing it via ZeroMe, I didn't get a response but my limit was raised. To 200KB.

So I guess I'll be having to do this again in another few months :)

Read more

Experimenting with alternatives to GIFs, part 2

on Mar 20, 2019 · 2 min read ·
1 comment

tag: 0net newb upload video images

In Part 1, I posted a brief experiment making and comparing (the file size at least) alternatives to animated GIFs. Part 2 is about getting these image & video formats to display on a ZeroBlog, and to do so in a way just like animated GIFs, i.e.

  • Displays automatically
  • Plays automatically
  • Loops automatically
  • No control bar obscuring part of the image

Because this blog here is not a regular ZeroBlog, I cloned ZeroBlog to create a sandbox for experimenting. It's here:


I'll summarize the results so far here.

Key:

  • WYSIWYG: Using ZeroBlog's WYSIWYG editing mode to insert the file into the blog post (ZeroBlog's editing mode automatically taking care of copying the file to the right folder and then inserting into the blog post the code to display it).
  • Manual: The blogger manually copying the file to the right folder and then manually writing and inserting the code (either Markdown or HTML) to display it.

Animated GIF (1.97MB)

  • WYSIWYG: Failure. Converted to still JPG.
  • Manual: Success. Using markdown code.
  • View post

Animated PNG (1.57MB)

  • WYSIWYG: Failure. Converted to still JPG.
  • Manual: Success. Using markdown code.
  • View post

Animated WebP (280KB)

  • WYSIWYG: Failure. Converted to still WebP.
  • Manual: Success. Using markdown code.
  • View post

MP4 (324KB)

  • WYSIWYG: No "Insert video" option.
  • Manual: Success... apparently... at first. Upon reload, Failure. It displays a still image. In order to play, the user has to right-click on the still image, select "Show controls" and then press play. Using HTML code.
  • View post

WebM (113KB)

  • WYSIWYG: No "Insert video" option.
  • Manual: Failure. Using HTML code. Just a big empty space. The HTML code keeps getting modified somehow.
  • View post

Conclusion

So while WebM is the smallest file size, I can't make it work on a ZeroBlog. MP4 appears to work at first, but upon reload is a failure (on 3 out of 4 criteria).

Also, even if I could get videos working on ZeroBlog, they might get blocked by browser extensions or settings that prevent autoplaying video.

So at the moment, I'm favouring lossy WebP, which:

  • works (manually at least)
  • gives a small file size (smaller than MP4)
  • can be inserted with markdown code (rather than HTML)
  • won't be blocked by video autoplay blocking methods

(I might also try out posting these formats on other ZeroNet platforms, e.g. MyZeroBlog {like this blog here} or even cloning GifTime. After all, it's got "Gif" in the name, and it publishes mp4s).

Read more

Experimenting with alternatives to GIFs, part 1

on Mar 16, 2019 ·
3 comments

tag: newb 0net images upload

As we all know, GIFs are evil. :)

Apart from issues like frame rate and limited colours, the main problem for us ZeroNet users is size: animated GIFs can easily have file sizes measuring in multiple megabytes. That's a problem when ZeroNet is based on ordinary users giving up their own disk space, bandwidth and data allowances to host and share everyone's zites.

But animated GIFs are fun!!! They automatically display, and move, and loop. All with no annoying sound! How to have all the fun of animated GIFs without all the bloat?

  • Here in part 1, I'll briefly check out some alternatives to GIFs, and how to make them. In part 2, I'll discuss trying to get these moving image formats to display in a blog on ZeroNet.

The GIFs

I recently made a post on my ZeroBlog "klu9's media share" about a comedy movie. A great opportunity for some funny moving images!

I found a couple of animated GIFs online I wanted to include in my post. But the file sizes were rather large: one was 543KB, the other 1.97MB. With only a few more posts like that, my blog would very quickly use up its default 10MB of space allowed on other users' hard drives and would start nagging people with "Set Limit to {X]MB" messages.

Here are the original GIFs, shared on ZeroUp Unlimited to avoid using up my zite's 10MB.

The Alternatives

Over the years, I'd heard of alternatives to GIFs, mainly video formats like .mp4 and .webm. (After all, what is an animated GIF but a form of short video?) Online image hosts like imgur and giphy automatically convert animated GIFs to video formats to save space and data traffic.

There are also formats originally designed for still images but which also have animated forms too: PNG and WebP.

First off, I thought I'd try converting the desired GIFs to WebM. There are command-line tools that can do this kind of conversion, like the Swiss Army Knife of video that is FFmpeg. But I'm a bit command-line phobic so I searched for a more GUI-fied option.

There are plenty of online converters, where you upload the GIF to a website, it converts it and then you download/save it back to your hard drive. That's probably the easiest option for just one or two GIFs.

There are also offline programs you can use on your computer, which often have the ability to convert whole batches of files. As I have a bunch of GIFs saved on my hard drive that I eventually want to convert, I've decided to investigate both options.

WebM

Some DuckDuckGoing led me to WebMConverter, a.k.a. WebM for Retards. Surely that'll be simple enough even for me! I downloaded, unzipped and ran it, at which point it told me it needed some other software called Avisynth and promptly opened a webpage to download it. After getting lost trying to find a 64-bit installer, I just went with the first link for 32-bit.

Once Avisynth was installed, I could run WebMConverter. I chose the 543KB GIF (monty.death.gif) as the Input file.

There are a few tabs and a heck of a lot of options available in WebMConverter but I only changed one setting: in the Encoding tab > General, I changed from Constant to Variable. I did this because of the easy-to-understand explanations.

Constant - This will make your video have a specific file size, and suffer low quality to match that size.
Variable - This will make your video get the size it deserves to look good.

WebMConverter
Full size

Once that was done, I clicked Convert. In just a second or two the job was done and I clicked WebMConverter's Play button to see the result. It was good. Then I went to check the file size... 16KB! Whoa!!! It reduced the file size by a factor of 33!!!

Here's the result on ZeroUp Unlimited:

I did the same with the 1.97MB GIF (every.sperm.is.sacred.gif), a more complicated image, and WebMConverter produced a video that was only 113KB. A reduction by a factor of only about 20; still great!

The Others: MP4, PNG, WebP

Later I went ahead and also converted that larger GIF to MP4, animated PNG, and animated WebP for comparison.

  • For MP4, I uploaded the GIF to giphy. Once the GIF's hosting page was displayed, I clicked on "Media" to the right of the image, which led to an MP4 page. (For some reason I couldn't do anything with it there but I was able to right-click on it, open in a new tab and then save it.)
  • For animated PNG, I used a GUI tool called gif2apng_gui.
  • For animated WebP, I used an online converter, ezgif.com and chose the option "Lossy" in order to get a smaller file size.

The Results

Here are all five versions of the larger file (every.sperm.is.sacred) shared on ZeroUp Unlimited:

Remember when comparing the quality:

  • I am not an expert on any of these formats or converting between them. I did little or no tweaking of options.
  • The starting point were GIFs I found online. You'd probably get better results starting from the original video.

The Conclusion

So it looks like WebM is the best alternative to animated GIF, at least when it comes to file size.

In part 2, I'll look at the next step: getting the image/video into a blog post.

Read more

This is not a ZeroBlog

on Mar 12, 2019
tag: 0net newb blog

When I first got started on ZeroNet, I saw quite a few "new and improved" claims about zites (better, more features etc). At that point I cloned MyZeroBlog, based on such claims.

But then after a few more days I realized a lot of claims were unsupported by evidence or just plain out of date. So I wanted to see if it was possible to change to regular ZeroBlog, as it was clearly receiving regular maintenance and updates. I couldn't really tell if anyone had improved MyZeroBlog since 2016.

I started a topic on ZeroTalk, I cloned one blog, can I change to a different one?, and followed some advice there.

I thought it had worked, basically because the blog didn't die and the two formats are so similar. But now that I've started another blog, cloned directly from ZeroBlog, I realize whatever it was I tried did not change this blog here from MyZeroBlog to regular ZeroBlog. I now realize it's still a MyZeroBlog.

And that explains why, when I'm editing a post, I don't get the blue circle giving a WYSIWYG editor & functions that @Geekless: mentioned, allowing image upload and enlarging, an issue I lamented in my post Experimenting with uploads: images. I have to do it all manually, dropping files into the /img folder and then making a link for the full size image.

On the other hand, MyZeroBlog does allow tagging. There are probably other differences too.

Now that I'm invested in this blog here (my bad), does anyone know of a way to add the WYSIWYG editor to a MyZeroBlog? Or at least the image upload and enlargement function?

Any help appreciated.

Read more

Build a database of media metadata on ZeroNet

on Mar 07, 2019 · 9 min read ·
2 comments

tag: 0net filesharing p2p

I'm posting here an unfinished draft of an article that got way too big and for which I was way underqualified. But the issue came up in a ZeroTalk thread:

That thread gives the gist of it, and my post was becoming too much about idea #2 (importing Bitzi Bitpedia database, which I realized probably wasn't useful, what with most people nowadays preferring high definition x264/x265 encodes rather than DivX .avi's). So as I'm probably never going to get round to completing this article (research, corrections, screenshots, links, cutting it way down), I'll throw the draft in here in case anyone's interested.

I've put the unfinished bits in code style.

Read more

Interview with Rabble re Twitter, IndyMedia, Secure Scuttlebutt

on Mar 04, 2019 · 2 min read ·
4 comments

tag: p2p SSB decentralization federation socialmedia tech

Here's a two-part interview with Evan Henshaw-Plath (a.k.a. Rabble, a.k.a. "Twitter employee #1") and who is now involved with the decentralized platform (/ZeroNet competitor?) Secure Scuttlebutt.

It's from Steal This Show, a podcast about P2P, decentralization, copyright issues etc, made by Jamie King, the director of Steal This Film about The Pirate Bay. Steal This Show is released via filesharing news site TorrentFreak

Steal This Show - 2019-02-05 - Software will Eat the World, with @Rabble - Part 1

Steal This Show - 2019-03-01 - Software will Eat the World, with @Rabble - Part 2

Description:

a two-part interview with @Rabble (Evan Henshaw-Plath) — activist technologist, co-founder of IndyMedia, and one of the originators of Twitter. We discuss the origins of Twitter in the protest organisation tool TxtMob; Evan’s work developing Indymedia and the early days of tech’s interaction with activism; how social media is continuing to mutate politics, for better and worse; how the sorting algorithms developed by Big Social are becoming indelibly embedded in our world — and finally Evan introduces the subject of part two: Silicon Valley’s hidden mission to restructure the world’s institutions via software.

Read more
Add new post

Title

21 hours ago · 2 min read ·
3 comments

tag:
Body
Read more

Title

21 hours ago · 2 min read
tag:

0 Comments:

user_name1 day ago
Reply
Body
<< 1 >>
This page is a snapshot of ZeroNet. Start your own ZeroNet for complete experience. Learn More