podbrushkin

joined 9 months ago
[–] podbrushkin@mander.xyz 2 points 1 month ago

I suspect there were no issue in the first place. A lot of time ago, if torrent client encountered files are missing, you had to run long process of verification for them to become available for seeding.

Now, if torrents switched to "files are missing" you can restart torrent client and seed like if nothing happened. No need for categories even. Also, function "Set location" can be used to move torrent to external storage.

[–] podbrushkin@mander.xyz 1 points 1 month ago

Confirmed as well. Also, stopping them before unmounting drive doesn't prevent missing files error.

 

OGDF C++ library provides some graph generators, I made it possible to use them from terminal, not by writing C++ client code. Got some pleasant pictures, here they are.

  1. Graphs of ~32 nodes, FMMM layout.
  2. Graphs of ~512 nodes, FMMM layout.
  3. Graphs of ~512 nodes, ForceAtlas2 layout.

Each node is labeled by it's degree - count of incoming/outgoing edges. Colored accordingly, from blue to red.

Powershell code for creating these pictures goes like that:

# define params for each generator
$genLines = @'
complement --n 32 --directed false --allowSelfLoops false
regularLatticeGraph --n 32 --k 4
regularTree --n 32 --children 2
completeGraph --n 32
completeBipartiteGraph --n 16 --m 16
wheelGraph --n 32
cubeGraph --n 5
globeGraph --meridians 8 --latitudes 4
suspension --s 32
gridGraph --n 8 --m 4 --loopN false --loopM false
petersenGraph --n 16 --m 2
emptyGraph --nodes 32
randomRegularGraph --n 32 --d 4
randomGraph --n 32 --m 100
randomSimpleGraph --n 32 --m 100
randomSimpleGraphByProbability --n 32 --pEdge 0.2
randomSimpleConnectedGraph --n 32 --m 50
randomBiconnectedGraph --n 32 --m 60
randomPlanarConnectedGraph --n 32 --m 60
randomPlanarBiconnectedGraph --n 32 --m 60 --multiEdges false
randomPlanarBiconnectedDigraph --n 32 --m 60 --p 0.0 --multiEdges false
randomUpwardPlanarBiconnectedDigraph --n 32 --m 60
randomPlanarCNBGraph --n 32 --m 60 --b 5
randomTriconnectedGraph --n 32 --p1 0.5 --p2 0.5
randomPlanarTriconnectedGraph --n 32 --m 80
randomPlanarTriconnectedGraph --n 32 --p1 0.3 --p2 0.3
randomTree --n 32
randomTree --n 32 --maxDeg 4 --maxWidth 10
randomDigraph --n 32 --p 0.1
randomSeriesParallelDAG --edges 60 --p 0.5 --flt 0.0
randomGeometricCubeGraph --nodes 32 --threshold 0.3 --dimension 2
randomWaxmanGraph --nodes 32 --alpha 0.5 --beta 0.2 --width 1.0 --height 1.0
preferentialAttachmentGraph --nodes 32 --minDegree 2
randomWattsStrogatzGraph --n 32 --k 4 --probability 0.3
randomHierarchy --n 32 --m 3 --planar true --singleSource true --longEdges false
pruneEdges --max_edges 60 --min_deg 2
'@

# generate graphs and layout/render with GephiCommander
$genLines | % {
$genCode = $_ -replace '\s+'
$outFile = Join-Path (gi ./img/) "$genCode$(Get-Date -Format FileDateTime).png"
$generatorArgs = $_ -split ' '
$expr = /bin/echo `& $ogdfCli --generate @generatorArgs --out-file delme.dot
Invoke-Expression $expr
@(
  @{op='import'; file='delme.dot' }
  @{op='preview'; 'background-color'='#171717'; 'nodeLabelShow'=$true; }
  @{op='colorNodesBy';column='degree'; mode='ranking'}
  @{op='labelNodesBy';column='degree'; }
  @{op='layouts'; values=@(
    @{name='ForceAtlas2'; steps=200}
    @{name='Expansion'; 'Density'=5000; steps=1;  }
  )}
  @{op='export';file=$outFile; timestamp=$false; resolution=@(512,512); }
) | ConvertTo-Json -d 9 | java -jar $gephiCommander -
$cmdEscaped = $generatorArgs -join ' '
magick $outFile -gravity SouthWest -undercolor "rgba(0,0,0,0.5)" -fill white -pointsize "%[fx:int(h*0.04)]" -annotate +10+10 $cmdEscaped $outFile
}

# Put all pngs into $files var and combine them into a grid
magick montage $files -tile "${cols}x${rows}" -geometry "400x400+2+2" -background "black" "img/grid_output$(Get-Date -Format FileDateTime).png"
[–] podbrushkin@mander.xyz 1 points 2 months ago

Seems reasonable, I didn't thought about port conflict. But in this scenario not running both clients at the same time also is an acceptable solution.

[–] podbrushkin@mander.xyz 3 points 2 months ago (2 children)

Thank you for this clarification, I will try to exit qbit and launch it again when HDD is accessible. Maybe it will reset those error statuses. I think it didn't work some years ago (error status remained, had to start verification).

[–] podbrushkin@mander.xyz 2 points 2 months ago (5 children)

"regular check" - do you mean verification? Verification takes a lot of time, verifying 10s of gigs will take a lot of time. What do you forget, stopping torrents?

[–] podbrushkin@mander.xyz 4 points 2 months ago (8 children)

I can move, but next time when I will launch qbt without external HDD accessible, those torrents will gain error status "Files not found" and will require a long verification process. This is what I'm trying to avoid.

I will check how categories can help managing this, thx

 

My torrent client downloads files to internal SSD. When I'm low on storage (often), I remove some torrents from qbittorrent and move files to external HDD. It means I can't seed them anymore.

I want to be able to seed HDD torrents periodically. What workflow can I employ? I can think of:

  1. Use 2 different torrent clients. One with "external" torrents only. It will never be launched when external HDD is not plugged in, therefore old torrents will not be marked as "missing files" and will not require verification every now and then;

  2. Use single torrent client which can accept CLI commands to start/stop specific torrents. Before detaching external HDD I will run a script to stop all external torrents. I will resume them back again only when HDD is accessible.

Any other ideas?

[–] podbrushkin@mander.xyz 2 points 2 months ago

You should’ve seen this place when maga activist was killed last year. Also can’t see none of your images except last two, lol.

 
  1. Power on your node in the morning. It is not connected to any smartphone or PC via BT/USB/HTTP;
  2. In the evening, connect to it via BT/USB/HTTP;
  3. Access all messages received in Primary channel and/or DMs.

Is it possible? How many messages does a node store in it's internal memory? How to access them via terminal? If node is restarted, are they erased?

I have a Heltec V3. In Meshtatsic CLI there's --listen option, but it's about receiving new packages in real time.

[–] podbrushkin@mander.xyz 1 points 2 months ago

I wonder what metrics provided by fediverse are actually reliable. E.g. views count on PeerTube videos are not, they can be adjusted by instance admin to his liking, lol.

[–] podbrushkin@mander.xyz 2 points 2 months ago

Nothing. All points randomly placed on a 2D plane, then forces are applied to them - all points are trying to move away from each other, but if 2 points are connected with an edge - they are attracted to each other. Check out this interactive demo of force-directed graph

[–] podbrushkin@mander.xyz 9 points 2 months ago (1 children)

Try image with white legend in bottom left corner.

 

cross-posted from: https://mander.xyz/post/49700074

On Mastodon, if you have an account on instance X, you can follow someone who is on instance Y. It creates a connection: X -> Y. If there are a lot of such follows, weight of this edge will increase, attractive force between points will be higher.

Original explanation on the page of Kaggle dataset:

"active users" graphs: For each instance, we consider the set of the 10K most recently active users. Then, for each user of an instance X, we consider the list of the users they follow, and add 1 to the edge from X to Y where Y is the instance the followed users. The weight of the edge from X to Y thus encodes how much the content seen on instance X is generated in instance Y. Note that this graph thus contains self loops.

I've tried to layout this dataset in Gephi, but it was a classic hairy ball - everyone is connected to everyone, amount of edges is too high comparing to number of nodes. Then, I've filtered out all EN instances and suddenly got a meaningful picture:

graph

What can we see? If English-speaking instances are ignored, German, French and Japanese languages are most common across Mastodon. Japan and Korea don't hang around much with other folks, while French, German and Spanish instances are quite interconnected between each other.

Size of nodes depends on centrality, post about centrality of Peertube instances is here.

Gephi table

Same, but Fruchterman-Reingold algorithm instead of ForceAtlas 2:

FR

Mastodon active users dataset can be downloaded here: https://www.kaggle.com/datasets/marcdamie/fediverse-graph-dataset-reduced

 

On Mastodon, if you have an account on instance X, you can follow someone who is on instance Y. It creates a connection: X -> Y. If there are a lot of such follows, weight of this edge will increase, attractive force between points will be higher.

Original explanation on the page of Kaggle dataset:

"active users" graphs: For each instance, we consider the set of the 10K most recently active users. Then, for each user of an instance X, we consider the list of the users they follow, and add 1 to the edge from X to Y where Y is the instance the followed users. The weight of the edge from X to Y thus encodes how much the content seen on instance X is generated in instance Y. Note that this graph thus contains self loops.

I've tried to layout this dataset in Gephi, but it was a classic hairy ball - everyone is connected to everyone, amount of edges is too high comparing to number of nodes. Then, I've filtered out all EN instances and suddenly got a meaningful picture:

graph

What can we see? If English-speaking instances are ignored, German, French and Japanese languages are most common across Mastodon. Japan and Korea don't hang around much with other folks, while French, German and Spanish instances are quite interconnected between each other.

Size of nodes depends on centrality, post about centrality of Peertube instances is here.

Gephi table

Same, but Fruchterman-Reingold algorithm instead of ForceAtlas 2:

FR

Mastodon active users dataset can be downloaded here: https://www.kaggle.com/datasets/marcdamie/fediverse-graph-dataset-reduced

 

cross-posted from: https://mander.xyz/post/49566880

This is a graph of Peertube instances following each other. There are 942 nodes and 10067 edges.

Eigenvector centrality

On Peertube, an instance X can follow an instance Y to let its users see all the videos posted on Y. This graph is a directed graph.

Color and size of nodes depends on how big their Eigenvector centrality is. Nodes which have 0 centrality are blue and small, nodes with bigger centrality are big and red.

What centrality represents? Instances which are not followed by anyone have 0 centrality. Instances (A) with a lot of followers (B) have bigger centrality. If those followers (B) themselves have followers (C), it means centrality of A will be even higher.

Does it mean anything in context of Peertube? I'm not sure. Considering chain of three instances: (A) <- (B) <- (C), when (A) posts a video, does it appear in (C)? Probably not. But if it was so, then centrality would've mean this: Videos posted on instances with high centrality spread across entire network, while videos posted on instances with 0 centrality are not visible anywhere else.

Here are top 10 instances and their centrality:

How to repeat this graph visualization

  1. Download latest Peertube instances.csv and interactions.csv files here: https://www.kaggle.com/datasets/marcdamie/fediverse-graph-dataset-reduced
  2. Import them to Gephi;
  3. Apply Giant Component filter to remove nodes which are not connected to biggest network;
  4. Apply ForceAtlas 2 layout;
  5. Run Eigenvector centrality Statistics (directed). It will add a new column to nodes table;
  6. Apply Nodes - Color - Ranking - Eigenvector centrality;
  7. Apply Nodes - Size - Ranking - Eigenvector centrality;
  8. Configure Preview and export.

Gephi

P.S. On colorful image used as thumbnail of this post nodes are colored by Modularity (community detection).

 

This is a graph of Peertube instances following each other. There are 942 nodes and 10067 edges.

Eigenvector centrality

On Peertube, an instance X can follow an instance Y to let its users see all the videos posted on Y. This graph is a directed graph.

Color and size of nodes depends on how big their Eigenvector centrality is. Nodes which have 0 centrality are blue and small, nodes with bigger centrality are big and red.

What centrality represents? Instances which are not followed by anyone have 0 centrality. Instances (A) with a lot of followers (B) have bigger centrality. If those followers (B) themselves have followers (C), it means centrality of A will be even higher.

Does it mean anything in context of Peertube? I'm not sure. Considering chain of three instances: (A) <- (B) <- (C), when (A) posts a video, does it appear in (C)? Probably not. But if it was so, then centrality would've mean this: Videos posted on instances with high centrality spread across entire network, while videos posted on instances with 0 centrality are not visible anywhere else.

Here are top 10 instances and their centrality:

How to repeat this graph visualization

  1. Download latest Peertube instances.csv and interactions.csv files here: https://www.kaggle.com/datasets/marcdamie/fediverse-graph-dataset-reduced
  2. Import them to Gephi;
  3. Apply Giant Component filter to remove nodes which are not connected to biggest network;
  4. Apply ForceAtlas 2 layout;
  5. Run Eigenvector centrality Statistics (directed). It will add a new column to nodes table;
  6. Apply Nodes - Color - Ranking - Eigenvector centrality;
  7. Apply Nodes - Size - Ranking - Eigenvector centrality;
  8. Configure Preview and export.

Gephi

P.S. On colorful image used as thumbnail of this post nodes are colored by Modularity (community detection).

 

Every single tracker displays how many seeders/peers each torrent have, but not ABB. Why?

Imagine being able to sort search results by seeders count. Or getting a table of top 200 audiobooks. Or finding valuable torrents which are on the verge of dying and supporting them.

In source of "Recent audiobooks" HTML page there's a trace of hidden "Seeds" column, maybe some time before it was possible to see these counts.

Are there any other way to get this data?

I've seen a torrent tracker which mirrors all torrents from ABB (adds them to its own catalog), seeds counts have been available for these torrents, but I don't remember what was this tracker, also it wasn't possible to find all torrents by source (ABB in this case).

EDIT: Probably I was talking about 1337x tracker, it does have some overlapping content with ABB, but it doesn't mirror everything. There's a table of all audiobooks which can be sorted by seeds, but you can't search through this category only.

[–] podbrushkin@mander.xyz 0 points 2 months ago

wdym 160+ likes, 40+ comments in a day for Redwall series in this community? Where’s my daily link to random title from Anna’s Archive or at least something about CIA, Stalin or US politics?

 

Doesn't seem like I'm doing much keeping this files alive on my PC.

no uploads

Can it be related to me not having public IP address? Or living in a place where not many people speak english (language of audiobooks)?

If connecting to me is troublesome because of dynamic IP, then why there's green globe at the bottom, what does it mean? Initially it's a fire icon. I did some port forwarding in wifi router settings, but I'm not sure if it have any meaning when you don't have public IP.

 

I've just created an account and in no more than two minutes a huge vagina appeared on my screen. This is problem.

Not that I have something against vaginas on my screen, but usually it takes some effort and commitment to make them available for observation. This time it was...

Well, ok, I did switch to Global feed, and there was a sensitive content warning, and I did click "See post", but still it was too fast. It was a huge blurred square on my screen, obviously I had to unveil this mystery.

Now I can't recommend Pixelfed to anyone in real life, because these pictures are just a few clicks away.

How I think it should be: when you create an account, "Hide NSFW content" option is enabled by default and such posts do not appear in your feeds or search results. If you wish, you can go to settings and enable it. "Blur NSFW content" is another checkbox enabled by default.

Right now there are no such settings at all. All I've found - is a quest of just six simple steps of creating a new filter. A quest should be needed to enable these images, not to disable them.

82
submitted 2 months ago* (last edited 2 months ago) by podbrushkin@mander.xyz to c/fediverse@lemmy.world
 

After all, why not? Why shouldn't I post it here?

It's a list of lists of Fediverse instances. Also, it contains some other lists of lists like this one. And a couple of miscellaneous resources. Tor tips can be ignored, they are region-specific.




Originally, I was trying to find a dataset, where for each instance there's a list of other instances it blocked/federated/defederated. In the process of trying to find this data I've compiled this list. Do you happen to know where I can find such data?

 

I've put a tiny piece of blue tape on Blu-ray disc with data, and, surprisingly, it was still fully readable (reading speed dropped at this place, that's all).

tiny tape

Next, I've put a slightly bigger piece of tape on it, and it made ~15% of disc unreadable. Reading "damaged" areas is quite slow, so I interrupted ddrescue/dvdisaster when data reading speed dropped too low, and it was around 15%.

big tape

Both results are strange to me. First is almost too good to believe, second is too bad. Maybe data behind little piece of tape have been recovered by internal Error Correction Code? And big piece, sure it didn't cover 15% of readable surface. Maybe full "recover" by ddrescue would have copied a lot more data, but I doubt it.

Why do you put tape on your disc?

I was trying to proof DVDisaster application works and will be able to recover missing data. I've failed to do so.

view more: next ›