I think it will come back in a great many forms via client apps though. So perhaps a demo app of that might be an idea?
I think it’s safe to think it will exist for sure though. We are not far away from ipfs naming and could use theirs even?
I am a bit removed and deep in knowledge transfer instead of data itself, but I think DNS like solutions will be aplenty
Now that is encouraging - It would prob be best to go read up on ipfs.
This one will be good to see if we are downloading efficiently i.e. like bittorrent and random chunks in parallel
Feck medium, we have a topic…
This screenshot suggests the upload speed has dropped dramatically, Seems my cloud instance only managed ~500 chunks in the last 3 hours.
Are others seeing anything similar ?
Are you being throttled for kicking the posterior out of your instances ?
Doesn’t seem that way to me.
I am seeing a slow down as well on a 40gb upload @Southside
here is the numbers from the upload and a ruff chunks per min calculation
⠁ [16:24:08] [#####>---------------] 16085/117090 16.3 chunks per min
⠲ [17:04:09] [#####>---------------] 16403/117090 7.9 chunks per min
⠈ [19:35:49] [#####>---------------] 17000/117090 3.9 chunks per min
One thing I noticed during the testnet tests using droplets is that seems the upload speed will be changing depending on the social working patten.
For the 2.4GB file I used for test, uploading time varies from 40mins to 2.3 hours.
And seems the night (GMT time), being the slowest, and Saturday and Sunday morning time (GMT time) being the fastest.
Not sure if it’s the same case here.
are the hashes of files remain same for those different copies ?
I am wondering maybe the 8 diff chunks is due to different name got used?
i.e. meta data of the file will affect a range of chunks due to the circular encryption.
@aatonnomicc, is that from home, or cloud?
For me, for fresh upload from home the speed seems as good as it used to be.
And @aatonnomicc, @Southside, @anon26713768, @shu, @storage_guy… and who else, who has many nodes up, how many nodes do you have?
I have now 15 nodes running from 6-7 years old laptop with 8GB of memory. I think it could run 2-3 times more, maybe even more. So I am wondering if this time, now that things are so stable, we could have this running from community machines only when Maidsafe pulls the plug?
Or if not now, how about the next version, and during holiday season? I also wonder how many nodes are needed for barely stable network now that these 2000+ nodes seems ultra stable? And what if most of the nodes are really scattered all over the Earth, with real-life latencies?
And @qi_ma, how many nodes we have now altogether? Is there a command for end user to find that information somehow?
And still one more thing for @qi_ma, @joshuef, @chriso …or anyone from the team:
Do we have a way to make an updateable list of files people have uploaded, that could be downloaded from the network? Registers somehow? It would be cool to have that central index where we could share our files inside the network, not just scattered in this thread.
Not at the moment. I do think something like that would be useful if it existed, but in my mind it probably wouldn’t be a core part of the network. It would perhaps be some kind of web app somebody else might setup in which people could voluntarily share links to public data.
Perhaps something like safe could generate output that would make such a thing easier for people to share.
In that case, if we have a list that grows very long, it would be quite expensive to re-upload it after a tiny addition? How about, what it was… versions, history, or some such, I don’t remember exactly?
What are those registers for? What are some practical use cases for them? (I don’t know anything about this stuff 
)
This could be a useful test of Safe Network’s CRDT/register functionality. I don’t know what state that is in atm, but in theory we can have a shared document based on a register that auto updates as different people edit it.
This requires someone to write a test app. This would also act as a good early demo for Safe Network, so could qualify for BGF funding if MaidSafe don’t themselves plan to create apps beyond the CLI.
I’m not aware that they’ve started to sort out the APIs yet, so this might be an area to get ready for rather than start coding.
Anyone interested in building this as an early Safe Network demo app? My hands are full but I would be happy to help in the background.
That I would love! 
Maybe I’m completely wrong about this, but what you’re talking about to me sounds analogous to something like a torrent tracking website. So it would just be a database of public links. You wouldn’t be adding to a new list each time.
Edit: actually, sorry, I see what you mean now. Using the network as the ‘database’. I’m not sure how it would be implemented.
Unfortunately, there is no single command to know how many nodes we have altogether.
But we can estimate it via the average peers among nodes.
I checked one of our droplet (running 20 nodes), and the avg is at 168, which indicates the overal number is around 3584.
Yeah, agreed an interesting feature to have.
However, not sure if there is privacy concerns for it.