Best RSS reader for someone who doesn't want their feeds tracked? (Including Firefox browser RSS add-on as well as stand-alone RSS)
I use stand-alone app FeedDemon. AFAIK app itself doesn't "track" you. Same can be hard to tell for sites you are getting articles from.
Unless you're authenticating, an rss feed is just a file download. Of course they will know IP address, which you can protect. It's my view that security and privacy is now impossible with browsing, and that exchange of structured messages (which rss is), is a much better approach because it separates acquisition, parsing, filtering and rendering processes - in a browser, those are all mashed together with the DOM plus Javascript, as well as material from 3rd parties. All because they want your eyeballs for advertising. At least with rss and edi, those things can happen without any particular rendering intent (which is where the privacy and security threats tend to come in any case). My ideal of computing is to spend as much time as possible away from the things, and to have various processes and bots humming away on my behalf which can far better be done with messaging approaches.
By "various processes and bots humming" you mean EDI and RSS or something else? What setup are you using?
I would avoid online services and use a local, standalone and open-source RSS feed reader such as the built-in one in Firefox, Liferea and QuiteRSS.
Exactly. But, apart from local RSS programs (including home-grown ones), I do not have an EDI solution. My view is that a more secure way of operating is to have message parsing and display clients under your complete control, backed by automation on the client which follow policies that you want to have happen. Not in the cloud. So, for example, if you express the wish to visit somewhere, it would send a message to an online marketplace (a bid), to solicit offers from vendors etc. with offers, based on some agreed tags in the messages. Those offers would be automatically processed and displayed. It does not need the cloud apart from a message exchange process and some form of hopefully distributed and privacy respecting directory. This is quite different to the eyeball-grabbing vendor aggrandisement we have at the moment. The problem of course is not technology but business model. We're in a phase at the moment where we have a dismal and consumer-exploitative business model. I think the only way to counter this is to have community-based initiatives which are genuinely community-enhancing. As a specific example, Uber has recently been refused a licence to continue operating in London. But this has nothing to do with the utility of the service, which many find useful. However, it's not at all clear to me that Uber has the interests of Londoners at heart - it pays (low) wages and few benefits to the drivers, and minimal corporate tax. How hard could it be for London to create an online community ride-sharing/taxi service for the benefit of the community? Obviously the answer to that is not in the scope of this forum, but for sure, there is nothing technically to stop privacy enhancing message-passing open standards to achieve great applications. Oh, and the other thing is to compel the social media offerings to have open apis which do not require their client and give them no visibility of content, and break up their control of marketplace and content. Wish me luck with that one.
I have similar view. Do you mean you developed/modified some? What do you use (language, library for language) to do this?
Mainly c# with standard .net libraries which directly support rss and atom, it's super easy - which leaves you free as a developer to do whatever filtering and UI you wish, as well as easy to meld to EF database data. Advantage being that you can obviously have it open source and custom. It's also pretty easy these days to directly parse xml schemas and process them.
1.How does one go about building this? Skill level required? How long would it take to attain that skill level? 2. Which language do you suggest and why that language over others to build this? 3. Directory is sitting on who's machine? 4.Is there a specialist forum that focuses on this subject? 5. I would would like to learn more about privacy enhancing message-passing applications, where should I start? 6. They don't do this because? Thanks
1.How does one go about building this? Skill level required? How long would it take to attain that skill level? I think the most important thing is to agree the schema between the parties. Of course, using rss or atom or edi standards helps, it depends what you're trying to communicate about. But, for example, I'd suggest we badly need, as consumers, the ability to transact at the structured and semi-structured level to exchange details of products and bids and offers, pretty much what EDI does in a complex way. The reason this hasn't happened is because the vendors and market-place providers do not want it to for the end user. It's not rocket science to have fields talking about weight, delivery, specifications of product etc. all in some form of structured and semi-structured way, so that automatic selection and filtering could be done by the CLIENT. And the reason for that is that a key element of what's needed is a distinct separation of data representation and visual rendering (if at all). That's what's wrong with the html5 dom, it's a horrid mixture of UI representational directives and what might be structured data, except site specific. 2. Which language do you suggest and why that language over others to build this? I don't think language is very important, except that it should be a modern high-level language with good IDE and library support, and protections for sillies like array bounds overflows etc. Plus built-in ASLR and so on. I'm happy with C# (but have been through assembly, Fortran, Basic and Java). I don't go in for language wars, the important thing is to have well supported libraries, and to avoid C or C++. I know you can use the latter "well", but there's a terrible legacy of unsafe code in C/C++. 3. Directory is sitting on who's machine? The directory would be the only thing sitting in the cloud, but instead of having very revealing email address handles etc, the handles for the user could just be unique anonymous tokens. There are ways now of doing encrypted database stuff that could allow some level of search without revealing that content. Obviously from a privacy perspective, the baseline is just a Guid handle identifier that is obtained side-channel. 4.Is there a specialist forum that focuses on this subject? 5. I would would like to learn more about privacy enhancing message-passing applications, where should I start? Not that I've found, other than here of course! It's my prediction that dark marketplaces will be the first exhibits of a move away from browser/web-based paradigm for interacting anonymously. The reason this is likely to happen is that web servers and browsers are way too complex and vulnerable. It's my view that community based standards should (eventually) be funded by national and local government - which they SHOULD do if they were looking after the people. For instance, I see no reason why a megalopolis should not insist on a standard/API for Uber-style ride hailing, based on messages of bids and offers specified in the standard. That way, no single company could dominate the marketplace, and the marketplace itself could easily be run by the community - it's just a reliable message exchange. 6. They don't do this because? They want your eyeballs, and they've given power to big business. The last thing they want is to cede control to automated services which give power to the client, and which inhibit marketing efforts and which break up networked monopolies. Look at the disastrous business model that the browser-web has become and you immediately see the problem. Back in the day, I enjoyed batch computing where computers did work on my behalf while I enjoyed the rest of my life. I do NOT want to be in front of a screen, wasting hours attempting to get information or locate suppliers. That's what the browser-terminal onto the web-server mainframe has become, and it's not in my interests.
Good thing about Java is programs written in Java are cross-platform, but it is more corporate specific. Recently Go and Rust languages are in great interests of FOSS community, but they don't have so much developed library ecosystem. Be aware in Rust memory is not automatically managed, so more burden is on programmer.
I agree. What I do is convert pages to atom feeds and filter these feeds with a blacklist, then send the updates as e-mail messages. That way I don't have to waste time with content I already know I don't want to see, as it is all filtered automatically. Plus some keywords are highlighted to make it easier to eyeball some categories of content. The process is similar when feeds don't exactly apply, like for price monitoring on Amazon. Extract the price, compare with the stored value, send an update e-mail if it decreased and is below the target. I can easily check 1000+ pages multiple times per day and not see a single advertising.
Just regular expressions right now, but I could plug-in anything available via command line (e.g. machine learning scripts).
Good work - how robust is this to changes in site layout, and how often does that tend to happen? Of course, all these marketplaces make this as difficult as they can, and it's doubtless against their Tos. Nevertheless. One thing I meant to mention above was the attraction of the Matrix messaging initiatives (used for example by Riot messaging, but capable of much more than that). It seems to me that programmatic exchange of structured messages using that as a substrate would provide a high-level base for no-ad-no-UI marketplaces (of course, the UI has to be there eventually, but can be in terms of digests or summaries or notifications and very much up to the client, NOT the merchant). https://matrix.org/
Everything else on the page can change as long as four specific strings and the relation among them stay the same. There has been no game of cat and mouse so far. There are ways to escalate the arms race. Never heard of matrix.org before, seems interesting. Will check it out later.
In agreement with Cohen in post #5... https://quiterss.org/ I've been using the portable version for several years, currently with 133 subscriptions. It's an open-source, cross-platform, stand-alone RSS/Atom client built with Qt-5.4.2, SQLite-3.23.1, WebKit-538.1. It's been in continued development and though the last update was May 31, 2018, it runs without issue on my Win7x64 desktop, checking for feed updates every ten minutes to three hours depending on the feed, an hour for most of 'em. The UI is an efficient, highly configurable management console. You can really bend this reader to your will. This one is a bit over the top IMHO, but it shows how far you can take it: https://screenshots.debian.net/screenshots/000/016/341/large.png There are tons of screenshots returned on a Google image search of quiterss. If you like big fancy fonts, wasted space, a locked layout and forced settings, QuiteRSS is not for you. It supports AdBlock filters where I use EasyList without element hiding rules, EasyPrivacy, Fanboy's Social Blocking List and Malware Domains, and a some of my own rules. I use the embedded browser, no images, no JavaScript, no background opening of links. Plaintext in the preview pane. You can configure to open a feed in your default browser with a right-click or a toolbar icon. I added an Other shield in MBAE as well.
