[transfer.sh] Default deletion behavior

The website states that default deletion is after 14 days. It would be better to keep files as long as they get downloaded frequently (e.g. 1 download every 30 days) as default behavior if user does not set a different deletion pattern.

@Alexey: I have created this topic in the non voting area for a reason as I don’t think it is suitable for voting.

1 Like

Are you willing to pay for more then 14 days?


I don’t think you can take money for it as there are probably hundreds of alternatives for free.
But you can make it an outstanding stable and feature rich service to promote the Tardigrade platform.
If it is your intention to turn this into a source of income, ad placement might be more promising.

1 Like

I think the point is that the suggestion would turn it from a 1 to 1 transfer service into a CDN like distribution service. We’re not talking a little more cost to swallow if that’s the case, but rather several orders of magnitude larger. SNOs still have to get paid and the service is offered for free. So yeah, there might be order models for income, but it would be much harder to swallow the additional costs if it could be used like that.


transfer.sh is open source. You can download it and run it on your machine. Would you allow me to store my files for more than 14 days for free? Don’t forget that you have to pay even for the 14 days usage. How do you make money out of that service?

The point is the 14 days limitation was not implemented by us. I hope the questions above are giving you an hint why the original developer has implemented this limitation. Even with that limit the “free” service was getting to expensive for him and he was forced to shut it down by his amazon bill. We would like to offer an cheaper alternative but I am not sure if we are willing or even allowed to take over the ownership of this project. At the moment we are keeping his project allive. Hopefully we can get his project back on track and he can make a profitable business out of it. I don’t like the idea of stealing ownership. As long as there is the possiblity to give him back full control over his project we should try to do exactly that.


It was my impression that transfer.sh has turned into a Storj project, as there has been much Storj involvement on the Github.
Recent PR activity has introduced the implementation of deletion feature so again it was my impression that it is Storjs responsibility.
If that’s not the case, that’s fine and there is no reason to give thought to this any longer.

1 Like

Surely it depends on what you want it to be. However still anyone can simply re-upload a file that has expired. In terms of cost this does not change a thing.

Ok, moved back. Sorry for misunderstanding the purpose

No problem. Thank you.

Hey @jammerdan! We are just providing transfer.sh the hosting capabilites and offset the HUGE costs that the owner had to carry while it was on AWS S3 Storage. While we contribute lots of code to the codebase, its mainly to suit our network optimally and kick out some missed parts of the implementation.

As you have spoken about it, currently transfer.sh does not have automatic file deletion as part of the app itself. The current setup involves another service to wipe the data after the 14 days expiry. As some of the previous responses already showed, indefinitely extending the storage comes at a very big cost. We have plans to make this use-case possible in the future and are in conversation with the owner and maintainer of the project about it. All i can say is: STAY TUNED


Thanks for clarification. I thought Storj was be running it as a showcase for the Tardigrade platform. I would do it. In any case it would serve better than filling nodes with dummy test data I suppose.
Are you interested in similar service? There is https://dfile.app/ with Git: https://github.com/coolcode/dfile
From their website they run on AWS S3 and should be facing the same issues like high cost like transfer.sh.
So maybe it would be worth to contact them either to pull their service over to Tardigrade or at least add Tardigrade support to their code in case someone else has use for it or forks it or something.