newsreader supplied by USENET provider limiting search results?

[AnotherBrian]

I'm able to access some old articles which cannot be found by using a usenet provider's newsreader. This makes me think that they are not making the entire newsfeed visible via their newsreader.

Is this so?

Related on this is the overall design of their service and their newsreader.

I'm guessing their client newsreader interfaces with a backend search engine on the provider's servers. Thus, the info made visible by the backend search engine limits what is visible via their client newsreader. While this may provide very fast access to search results, its disappointing if only a bit of what is out there is accessible.

in the old days this is how a newsreader worked. First, for the newsgroup of interest, the article headers had to be downloaded. Once those were on the client machine, the selected individual articles could be downloaded.
the header files could be huge both in storage volume and time to download and so for example fetched headers could be limited to a date range or since last fetched.
The behavior of the usenet provider's newsreader suggest they are not downloading the header files into the client machine.

Any suggestions on a good newsreader to use which would bypass the usenet provider newreaders limitations or maybe it is just a usage issue on my part? Klibido use to be a preferred binaries downloader but requires a build by the user these days. I don't know if there are any issues to getting it to work.

[MoebusNet]

Have you tried: https://binsearch.info/ ?

Free web-based search engine; monthly fee for advanced features like no-longer-current articles. Might be worth a look.

[AnotherBrian]

I think I figured it out. I'm thinking that the USENET provided reader screens material perceived as copyrighted. They didn't say this outright but that is the impression I got.

Another possibly is their backend search engine may not be picking up all the content. The design must be something like they have a database mapping search terms to header ids. In background mode they have to construct the database and then add to the database over time as new articles come in. Any of that may get corrupted and then it would be a large effort to rescan all that content. Also they would need audits to assure to detect something is broken.

Some of the features on the provider's newreader are nice such as taking a peek at content before downloading a large file.

I don't consider sabnzb as a USENET reader. It requires an nzb indexer and a significant amount of content is not based on nzb.

pan chokes on large sets of header files. klibido I don't think will run on ubuntu but perhaps on kubuntu. Not sure on this.

Are there any good newsreaders out there for linux?

Thank you for the suggestion about binsearch.info.

[sdsurfer]

There is also the Wayback Machine, the internet archive but not sure if it archives news feeds.

[AnotherBrian]

I ended up finding another newsreader, newsflash plus by ensisoft.
This newsreader is header file based meaning header files need to be downloaded before the individual articles can be read.
So it is like pan in this regard. It also is able to handle the large files found on usenet like rar files etc and unpack them.

where pan goes into high cpu usage when the number of downloaded header files is large, newsflash can handle it.

the newsreader supplied by newshosting, although is has nice preview features, the backend search function appears to greatly restrict the viewing of huge amounts of usenet.

I also subscribed to eweka to do some experiments between the two platforms (eweka and newshosting). I did full downloads of all available header files in several newsgroups using newsflash. it appeared to me that both eweka and newshosting have identical headerfiles in the several newsgroups.

That suggests that eweka (who advertises a greater retention period) and newshosting (who advertises a lesser retention period) are are providing identical retention periods and possibly using a common source for usenet content. If that is indeed the case, then there is no advantage between the two from a retention perspective that I can see.