posted by
NCommander
on Monday June 16 2014, @02:09PM
from the expecting-torchs-and-pitchforks dept.
NOTE: For those who aren't interested in "meta" articles, just ignore this one. There's an update coming down the pipe to allow people to filter out content from the main page which will hopefully be live rather soon.
As some have may gathered, given our recent push to collect statistics on the site, we're working hard to identify and understand viable fundraising methods that we can use to keep SoylentNews up and running without alienating the community. We are looking at a number of options, including subscriptions for premium services, and it remains our goal to avoid running ads on the site, if at all possible. As it turns out, getting a firm grasp on realistic fundraising estimates has proven to be the determining factor in how we incorporate.
As I've stated previously, I had intended for us to start by forming a not-for-profit (NFP) with the possibility of 501(c)(3) tax-exempt status, and I had several conversations with various legal and tax authorities about this. The idea was that the NFP would serve as an umbrella organization, under which independent sites, including SoylentNews, could operate. What I didn't realize at the time was that setting up a structure like this from the beginning would be far more difficult and expensive than I anticipated. So, simply put, we've found it necessary to change plans.
As described in more detail below, NFPs have heavier burdens due to the strict rules imposed on them for the usage of revenue and resources. Furthermore, in talking with our prospective lawyer and accountant, it quickly became evident that we would have significant burdens and hurdles to clear if we proceeded with the original plan to set up the NFP immediately. When I met face to face with the staff, we began working out what the costs of NFP incorporation was going to look like, and it was getting close to $8,000 USD simply to get everything vetted by the lawyer and accountant because of retainers and other such overhead; this was just to get things established and did not deal with items such as ongoing financial reporting and other requirements.
I always wanted SoylentNews (SN) and its staff to have relative autonomy as a not-for-profit, with the site and its community paramount, and to slowly morph to a fully democratic model once we were relatively stable. To do so would require a defined relationship between SoylentNews and the NFP in each organization's bylaws, which opened all sorts of legal questions on how funding and such would be handled. It quickly became evident that the cost of setting up this kind of structure would be very high, and it was unclear if SoylentNews could re-coup those costs in a timely matter. We could easily end up in a situation where resources were exhausted and yet we needed more help to get things established. It would be an unworkable situation, and not one that would bode well for the future of the site.
Here's a direct example: in the United States, NFPs need to be licensed to perform fundraising in each individual state they're receiving money from. As I outlined in both the guiding statement and the manifesto, one of the ways of raising revenue is to offer subscriptions. The money for those subscriptions would be used to support site operations. Would that be considered fundraising?
Our lawyer hi-lighted this issue when we proposed our revenue models, and wasn't sure, he said he'd have to forward the question to a certified public account or to the state for an answer. This is only one example of "known unknowns" we came up with (mrcoolbp filled two legal sized pages with these when we sat face to face). Needless to say, the situation wasn't looking good, but we've come up with a plan, and I want the community to vet it before we proceed.In general, companies (not NFPs) have a legal responsibility to raise funds and revenue for their owners, a concept known as "shareholder primacy", first defined in Dodge v. Ford Motor Company, while not-for-profits only exist to further their mission (in exchange for the possibility of eventually getting tax-exempt status), and are subject to strict and complicated rules which are often not clear cut. SoylentNews and its staff itself exist to serve its community and its mission - in spirit, we're already a not-for-profit - but at this moment we simply don't have the resources to avoid hitting landmines in what has proven to be very murky waters.
This left us in a difficult catch-22. We could proceed with what resources we had, and risk everything blowing up in our face or try to find a simpler way to get things set up in the short term to allow us to begin fundraising to keep the site running. Of course, I realized that if we were to about-face and incorporate as a for-profit, it would be both a slap in the face of the community and would risk destroying the rapport we built up with the community and well as any trust or good-will we've built up since go-live. Fortunately, we think we've found a way to proceed that will maintain the spirit of what I want SoylentNews to be.
During our face to face, matt_ offered an alternative that I had not previously heard of: a benefit corporation (also known as a B Corp). For those of you scratching your heads wondering what it is, you're not alone. Benefit corporations are a new type of corporation that came to exist in 2010. Summed up, a benefit corporation is a for-profit corporation that exists for public benefit and can be seen as a middle ground between a traditional for-profit entity and a not-for-profit. Under a B corporation, the board of directors are bound to "pursuing the creation of general public benefit, and any named specific public benefits, is considered to be in the best interests of the corporation." In line with this, our certificate of incorporation would contain the following statement, or one like it: The specific public benefit purpose of the corporation is to engage in and promote free, open journalism through the production and publication, and community-sourced analysis and discussion of news and original and third-party-sourced works of fact and opinion. Under a B corporation, as long as we succeed in this mission, we will have, by definition, fulfilled our corporate responsibility.
I realize this is a fairly large departure from what our original plan was, but I think it would be folly to charge ahead with the original plan in light of what we know now with the resources we have available. As a B corporation, we will be able to operate as a traditional corporation in terms of both raising and spending revenue and would be treading in much safer legal waters. Furthermore, our incorporation costs will be in the hundreds, not thousands, of dollars. Finally, and most critically, setting the site up in this way does not prevent us from establishing the not-for-profit in the future once we are financially stable. As described below, this plan allows us to turn into (technically, be acquired by) the not-for-profit, if and when doing so becomes financially possible.
I do, however, want to make this clear, right here, right now. I'm not going to do something that's going to outrage the community. If you think we're in error, let us know. If we have to, we'll scrap the B corporation, and figure out a way to make things work as a pure not-for-profit. I'll be damned before I piss off the community and cause a "beta-like" folly which alienates everyone -- I'm pretty sure all the staff would agree with that sentiment. The wikipedia page and the Benefit Corporation Information Center have considerably more detail; I ask that folks take a look at these pages to understand the details at hand before coming to a conclusion one way or the other...
Now, with all that said, I still feel that eventual incorporation as a not-for-profit could still be beneficial for us. In the revised incorporation plan, after we incorporate as a B corporation, we intend to get the corporate charter and operating procedures of the site straightened out to the point that SoylentNews is a self-sufficient and fully operational independent entity. After that process is complete, I intend to use the resources of the B corporation to re-evaluate becoming a NFP corporation. If we (both the staff and the site) feel this is beneficial once we are able to fully answer our major outstanding "known unknowns", we will sell the B corporation to the newly-incorporated not-for-profit, and the site will continue as a benefit corporation owned by a not-for-profit corporation.
The eventual end result will be an independent SoylentNews that is able to operate freely as an independent entity with manifest destiny, and which will be owned by a not-for-profit that will protect the site and fight for the rights and protections that we all believe in. I hope to build SoylentNews into a shining example of what journalism and press should be, with a structure that enables us to fight to protect our rights, increase public education, and help restore integrity to the field of journalism.
This discussion has been archived.
No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
(Score: 1, Interesting) by Anonymous Coward on Monday June 16 2014, @03:16PM
by Anonymous Coward
on Monday June 16 2014, @03:16PM (#55929)
> the reason so many sites out there are stuffed with ads is because it takes lots of money to run a popular site.
Here is a more radical proposition that is more about possibilities than anything that could be implemented tomorrow:
Re-architect the site to distribute the majority of operational costs to the users. Instead of hosting the entire site on a central server, distribute the load with something like bittorrent. We would still need a central server to handle synchronization of new stories, new comments and moderations but that seems like a small fraction of the resource consumption - most of which is just CPU and bandwidth costs to serve pages.
Imagine a design that breaks the system down into three types of "files":
(1) Front page stories (2) Comment Section for each story (3) List of moderations for each Comment Section
If each of those were treated as single semi-static files we could regenerate the file and push a new copy out via P2P whenever there was an update or once every 30 seconds (which ever is longer). With smart javascript the client could take care of filtering and displaying the comments according to each user's preferences.
It should be possible to do it 100% in the browser with HTML5 technologies like WebRTC. [google.com]
I think it dovetails nicely with the "soylent news is people" concept - not only would the content be us but the servers would be us too.
When you are done re-architecturing the Web completely, please come back to us with a detailed design, and if possible a working proof of concept. We'll then hire a 1000 strong team of monkeys to code it on typewriters.
Meanwhile, although a cool concept, not exactly helpful for what we've got now.
-Jar
-- This is my opinion, there are many others, but this one is mine.
(Score: 0) by Anonymous Coward on Monday June 16 2014, @03:43PM
by Anonymous Coward
on Monday June 16 2014, @03:43PM (#55956)
>> is more about possibilities than anything that could be implemented tomorrow: > > Meanwhile, although a cool concept, not exactly helpful for what we've got now.
You are welcome, I'm always glad to give a douche an opportunity to restate the obvious. Thanks for contributing to the discussion!
Skip his implementation details, set up UUCP cnews / usenet servers not connected to the greater usenet in any way, and containing a couple moderated newsgroups (with aggressive anti spam cancelbots and require valid GPG sig or the article gets zapped) and run with it. Its a bit of sysadmin type work more so than re-architecting stuff and writing code.
Distribute the whole works as Debian live CD and also a ras-pi image. So any old machine can Soly itself into being part of the network.
Put a web client of sorts on the server too. So burn a SD card for a ras-pi, set up its network and an upstream peer, and navigate over to the ras pi in your web browser and away you go. This might require a small amount of glue code to be written.
As for finding peers to connect to the network keep irc.soylentnews.org or just configure peer.soylentnews.org to allow anyone with an account and GPG sig to connect.
It was either usenet or some insane varient of fidonet, couldn't decide which.
The problem with IT is nothing is every really new. Describe fidonet in everything but name and ...
Believe it or not, this is actually on my long term TODO list (you actually even got right the design I wanted, with slashd writing the site index as a UUCP batch file, feeding into a local INN instance, mostly because I don't want to write the entirity of the NNTP protocol in Perl as a interface to the DB).
I had long talks with mechanicjay on how we could do this ...
Soylent is already available through TOR, and while your idea has some merit; you are promoting a distribution method that would only succeed if enough peole were willing to host and seed the daily files. - that's a tall order - and it would be a high barrier for new people to join.
(Score: 0) by Anonymous Coward on Monday June 16 2014, @05:31PM
by Anonymous Coward
on Monday June 16 2014, @05:31PM (#56015)
> Soylent is already available through TOR
TOR does not offload any server workload, it isn't P2P in anyway. I am actually posting through TOR right now.
> would only succeed if enough peole were willing to host and seed the daily files
I'm not proposing that there be daily files. I don't even know what you mean by that.
What I am proposing is that the role of the central server be reduced to the bare minimum such that anyone with an HTML5 browser can just hit the server and "it just works." If there are no other users than the central server still does all the work (which will be miniscule if there is only one user), but the more users there are the more the load gets picked up by those users.
Soylent is already available through TOR, and while your idea has some merit; you are promoting a distribution method that would only succeed if enough peole were willing to host and seed the daily files. - that's a tall order - and it would be a high barrier for new people to join.
If we ever get the infrastructure together for a distributed SN, I volunteer. I run my own domain, and I peak at about 1% capacity. If someone puts together a how-to with software packages, I'll gladly donate resources.
Soylent is already available through TOR, and while your idea has some merit; you are promoting a distribution method that would only succeed if enough peole were willing to host and seed the daily files. - that's a tall order - and it would be a high barrier for new people to join.
Well, he did suggest using WebRTC. If I understand it correctly, that would mean every single user browsing the site would also act as a server -- but only while they have the site open. That *might* actually work for a site like this -- it's not uncommon for me to leave the tab open for the majority of my work day, so I'd be serving the content for nearly eight hours a day. Then again, that might end up just getting you swiftly blocked from major corporate networks...
mm. i cant say I'm familiar with WebRTC; but I am sure that if my work PC was acting as a server that the site would be blocked in short order by the IT staff here. (all "streaming" services are blocked to start with; if it looks like I'm hosting, that could cost me my job on top of it)
Re-architect the site to distribute the majority of operational costs to the users. Instead of hosting the entire site on a central server, distribute the load with something like bittorrent.
An interesting idea, albeit quite radical thus coming with all the risks incurred by revolutions.
But, maybe a we can find a middle ground? Look, for example, if SN would be to offer an API to access all the posts and comments and, subject to size/frequency/other restrictions, allow posting comments, I'd be more willing to pay an extra subscription fee for a token that grants access to this API (especially if it would bypass the lame too-many-caps-is-yelling filter and would genuinely support UNICODE). I don't think the API can be very hard to come with, all GUI-related and UE stuff can be ignored: JSON is enough.
UTF-8 is mostly fixed on dev (some hangups in backslash). Check out http://dev.soylentnews.org/ [soylentnews.org] to see it in action. This update will land in the Nexuses update soon.
(Score: 0) by Anonymous Coward on Monday June 16 2014, @04:41PM
by Anonymous Coward
on Monday June 16 2014, @04:41PM (#55978)
> An interesting idea, albeit quite radical thus coming with all the risks incurred by revolutions. > > But, maybe a we can find a middle ground?
I like the way you are thinking.
The only radical part of the proposal is the P2P stuff. All the other parts could be implemented on a standard webserver. It could be a slow migration of segmenting the pages to work as described and implementing the client-side code to handle it.
Once that is all solidly in place you could start fiddling with the P2P stuff and probably run it in parallel with the server-centric system for as long as necessary, maybe even permanently as a fall-back for people who can't run the P2P due to browser or bandwidth limitations.
I'm not even original [ihackernews.com]. But I think my proposal may bring an extra support for SN (not only on the financial side), but by allowing an independent involvement in the development side of SN. Why independent is crucial? Speaking for myself: I didn't dare raise my hand on the call for devs, etc because my spare time is not predictable (so I can't commit) and I don't know PERL. But I'd love to start some pet-projects built around SN
maybe even permanently as a fall-back for people who can't run the P2P due to browser or bandwidth limitations.
This + I imagine we don't want to raise the barrier for new people joining SN by asking them: you either run part-of-a-server or don't join at all.
Thank you indeed.
Hate to pester you, but... can you at least think of building an API around SN? UTF8 and yelling-filter weren't actually the main reasons for my pretty-please asking an API.
What would you want in an API. If I ever have the time, I'd love to mirror the site onto an INN server (bidirectorially) and then perhaps even seed it into USENET which should cover 95% of what an API would cover.
An API able to mirror back and forth to an INN server will probably cover my needs too (especially if it would allow detecting and mirroring only "what's new"), but let's not forget the personal journals - most of the time, they aren't "visible" enough (i.e. if you don't look specificallyon the user page, one user at a time and one page at a time, you aren't going to see them).
If you like, I'd try to put together a high level spec over weekend.
Sure, though please realize that dev resources are somewhat limited, so I have no idea when this would come around (unless of course you'd like to code yourself :-))
(unless of course you'd like to code yourself :-))
I... sort-ish... might if you're OK with either PHP or Java riding on top of the database (the sort-ish: would be a pet project for me, as such, if I'm lucky I'll finish it in one larval stage [catb.org] sitting - at least the "read" part - or I might get dragged into the mundane and put it on hold).
I can't count how many times I've sat there *wishing* some site or service had an API. I've done quite a bit of PHP "screen-scraping" of sites to get at the data I wanted. Not sure what I'd use it for, but I've got a bit of a home theater and home automation system set up on my Raspberry Pi, and I'm sure I could find some way to incorporate Soylent into that ;) Would gladly pay a couple bucks a month if I got a simple API with clean JSON-encoded results! Maybe I'd do a 'today's headlines' alarm clock again. Although that just needs RSS really...
Hmm...what could I do with a full API like that....?
That would make it REALLY easy to make some nice mobile interfaces though, that would be great!
Hmm...what could I do with a full API like that....?
For my case? Say... I have a bunch of mod points that can't spend most of the time (by the time they expire).
How about something (like a fragment of an app) like: show me messages that I could mod? (don't show me +5-es first, that's no point for me to mod them up again. Don't show me messages in the stories that I posted: I can't mod them anyway).
Yes, such an app would require to know my identity. If this is done using a token, what SN is "selling" me is not a subscription, but the "token to access a functionality". No hosting, no email or some other ancillaries, but something that has an added value to the very service I'd be using, something that can be quantized in terms of costs - count the calls/measure the traffic - so that the operator can make his/her minds in terms of profitability, etc.
Your needs are good enough with the HTML face that SN dev-team is maintaining? Don't pay anything. You want more: write your app (and to access the data you need in a way that you like) or buy an already written app. Besides, the capability to tinker with the very things one likes should be the wettest dream of a nerd (in this case, of the programming kind).
Sounds neat. But it also sounds suspiciously like Usenet, which exists, and where you can already go to read and comment on tech stories. Start at comp.misc, which is covering many of the same stories being covered here.
The federated model also sounds somewhat like what is being built over at Pipedot. I don't understand it totally because Bryan hasn't totally spilled the beans on his vision, but it involves a network of federated sites, and I can only assume, some sort of transfer or share of messages or stories among them. It's marching forward, whatever it is!
Someone said, sooner or later, every web forum's new inventions bring it a step closer to Usenet. Here we go!
-- Dad always thought laughter was the best medicine, which I guess is why several of us died of tuberculosis - Jack Handey
(Score: 0) by Anonymous Coward on Monday June 16 2014, @05:16PM
by Anonymous Coward
on Monday June 16 2014, @05:16PM (#56007)
> Sounds neat. But it also sounds suspiciously like Usenet, which exists,
The key difference is that usenet isn't very dynamic. You can't set up your own NNTP server just by clicking a link. So you have to get a client, find a server, get an account, etc. Plus usenet has spam issues. Usenet might make a reasonable medium for a higher level "protocol" that uses crypto signatures to prevent spam (standard usenet moderation won't stand up to a smart attacker nowadays) but then you are looking at a custom usenet client - well you need that for handling moderations too. I see usenet as useful for archival but not so robust for offloading server costs to users in an automatic way.
(Score: 1, Interesting) by Anonymous Coward on Monday June 16 2014, @03:16PM
> the reason so many sites out there are stuffed with ads is because it takes lots of money to run a popular site.
Here is a more radical proposition that is more about possibilities than anything that could be implemented tomorrow:
Re-architect the site to distribute the majority of operational costs to the users. Instead of hosting the entire site on a central server, distribute the load with something like bittorrent. We would still need a central server to handle synchronization of new stories, new comments and moderations but that seems like a small fraction of the resource consumption - most of which is just CPU and bandwidth costs to serve pages.
Imagine a design that breaks the system down into three types of "files":
(1) Front page stories
(2) Comment Section for each story
(3) List of moderations for each Comment Section
If each of those were treated as single semi-static files we could regenerate the file and push a new copy out via P2P whenever there was an update or once every 30 seconds (which ever is longer). With smart javascript the client could take care of filtering and displaying the comments according to each user's preferences.
It should be possible to do it 100% in the browser with HTML5 technologies like WebRTC. [google.com]
I think it dovetails nicely with the "soylent news is people" concept - not only would the content be us but the servers would be us too.
(Score: 5, Funny) by Jaruzel on Monday June 16 2014, @03:21PM
When you are done re-architecturing the Web completely, please come back to us with a detailed design, and if possible a working proof of concept. We'll then hire a 1000 strong team of monkeys to code it on typewriters.
Meanwhile, although a cool concept, not exactly helpful for what we've got now.
-Jar
This is my opinion, there are many others, but this one is mine.
(Score: 0) by Anonymous Coward on Monday June 16 2014, @03:43PM
>> is more about possibilities than anything that could be implemented tomorrow:
>
> Meanwhile, although a cool concept, not exactly helpful for what we've got now.
You are welcome, I'm always glad to give a douche an opportunity to restate the obvious.
Thanks for contributing to the discussion!
(Score: 2) by VLM on Monday June 16 2014, @03:56PM
Skip his implementation details, set up UUCP cnews / usenet servers not connected to the greater usenet in any way, and containing a couple moderated newsgroups (with aggressive anti spam cancelbots and require valid GPG sig or the article gets zapped) and run with it. Its a bit of sysadmin type work more so than re-architecting stuff and writing code.
Distribute the whole works as Debian live CD and also a ras-pi image. So any old machine can Soly itself into being part of the network.
Put a web client of sorts on the server too. So burn a SD card for a ras-pi, set up its network and an upstream peer, and navigate over to the ras pi in your web browser and away you go. This might require a small amount of glue code to be written.
As for finding peers to connect to the network keep irc.soylentnews.org or just configure peer.soylentnews.org to allow anyone with an account and GPG sig to connect.
It was either usenet or some insane varient of fidonet, couldn't decide which.
The problem with IT is nothing is every really new. Describe fidonet in everything but name and ...
(Score: 2) by NCommander on Monday June 16 2014, @04:33PM
Believe it or not, this is actually on my long term TODO list (you actually even got right the design I wanted, with slashd writing the site index as a UUCP batch file, feeding into a local INN instance, mostly because I don't want to write the entirity of the NNTP protocol in Perl as a interface to the DB).
I had long talks with mechanicjay on how we could do this ...
Still always moving
(Score: 2) by Blackmoore on Monday June 16 2014, @04:17PM
Soylent is already available through TOR, and while your idea has some merit; you are promoting a distribution method that would only succeed if enough peole were willing to host and seed the daily files. - that's a tall order - and it would be a high barrier for new people to join.
(Score: 0) by Anonymous Coward on Monday June 16 2014, @05:31PM
> Soylent is already available through TOR
TOR does not offload any server workload, it isn't P2P in anyway. I am actually posting through TOR right now.
> would only succeed if enough peole were willing to host and seed the daily files
I'm not proposing that there be daily files. I don't even know what you mean by that.
What I am proposing is that the role of the central server be reduced to the bare minimum such that anyone with an HTML5 browser can just hit the server and "it just works." If there are no other users than the central server still does all the work (which will be miniscule if there is only one user), but the more users there are the more the load gets picked up by those users.
(Score: 1) by bziman on Monday June 16 2014, @11:52PM
If we ever get the infrastructure together for a distributed SN, I volunteer. I run my own domain, and I peak at about 1% capacity. If someone puts together a how-to with software packages, I'll gladly donate resources.
(Score: 2) by urza9814 on Tuesday June 17 2014, @01:52PM
Well, he did suggest using WebRTC. If I understand it correctly, that would mean every single user browsing the site would also act as a server -- but only while they have the site open. That *might* actually work for a site like this -- it's not uncommon for me to leave the tab open for the majority of my work day, so I'd be serving the content for nearly eight hours a day. Then again, that might end up just getting you swiftly blocked from major corporate networks...
(Score: 2) by Blackmoore on Tuesday June 17 2014, @02:08PM
mm. i cant say I'm familiar with WebRTC; but I am sure that if my work PC was acting as a server that the site would be blocked in short order by the IT staff here. (all "streaming" services are blocked to start with; if it looks like I'm hosting, that could cost me my job on top of it)
(Score: 2) by c0lo on Monday June 16 2014, @04:23PM
An interesting idea, albeit quite radical thus coming with all the risks incurred by revolutions.
But, maybe a we can find a middle ground?
Look, for example, if SN would be to offer an API to access all the posts and comments and, subject to size/frequency/other restrictions, allow posting comments, I'd be more willing to pay an extra subscription fee for a token that grants access to this API (especially if it would bypass the lame too-many-caps-is-yelling filter and would genuinely support UNICODE).
I don't think the API can be very hard to come with, all GUI-related and UE stuff can be ignored: JSON is enough.
https://www.youtube.com/watch?v=aoFiw2jMy-0 https://soylentnews.org/~MichaelDavidCrawford
(Score: 3, Informative) by NCommander on Monday June 16 2014, @04:34PM
UTF-8 is mostly fixed on dev (some hangups in backslash). Check out http://dev.soylentnews.org/ [soylentnews.org] to see it in action. This update will land in the Nexuses update soon.
Still always moving
(Score: 0) by Anonymous Coward on Monday June 16 2014, @04:41PM
> An interesting idea, albeit quite radical thus coming with all the risks incurred by revolutions.
>
> But, maybe a we can find a middle ground?
I like the way you are thinking.
The only radical part of the proposal is the P2P stuff. All the other parts could be implemented on a standard webserver. It could be a slow migration of segmenting the pages to work as described and implementing the client-side code to handle it.
Once that is all solidly in place you could start fiddling with the P2P stuff and probably run it in parallel with the server-centric system for as long as necessary, maybe even permanently as a fall-back for people who can't run the P2P due to browser or bandwidth limitations.
(Score: 2) by c0lo on Monday June 16 2014, @05:53PM
I'm not even original [ihackernews.com].
But I think my proposal may bring an extra support for SN (not only on the financial side), but by allowing an independent involvement in the development side of SN.
Why independent is crucial? Speaking for myself: I didn't dare raise my hand on the call for devs, etc because my spare time is not predictable (so I can't commit) and I don't know PERL. But I'd love to start some pet-projects built around SN
This + I imagine we don't want to raise the barrier for new people joining SN by asking them: you either run part-of-a-server or don't join at all.
https://www.youtube.com/watch?v=aoFiw2jMy-0 https://soylentnews.org/~MichaelDavidCrawford
(Score: 1) by paulej72 on Monday June 16 2014, @05:46PM
Also to make utf-8 work the too-many-caps-is-yelling filter has to go as well as the whitespace filter (some languages do not use whitespace).
Team Leader for SN Development
(Score: 2) by c0lo on Monday June 16 2014, @05:59PM
Hate to pester you, but... can you at least think of building an API around SN? UTF8 and yelling-filter weren't actually the main reasons for my pretty-please asking an API.
https://www.youtube.com/watch?v=aoFiw2jMy-0 https://soylentnews.org/~MichaelDavidCrawford
(Score: 2) by NCommander on Monday June 16 2014, @10:20PM
What would you want in an API. If I ever have the time, I'd love to mirror the site onto an INN server (bidirectorially) and then perhaps even seed it into USENET which should cover 95% of what an API would cover.
Still always moving
(Score: 2) by c0lo on Monday June 16 2014, @11:33PM
An API able to mirror back and forth to an INN server will probably cover my needs too (especially if it would allow detecting and mirroring only "what's new"), but let's not forget the personal journals - most of the time, they aren't "visible" enough (i.e. if you don't look specificallyon the user page, one user at a time and one page at a time, you aren't going to see them).
If you like, I'd try to put together a high level spec over weekend.
https://www.youtube.com/watch?v=aoFiw2jMy-0 https://soylentnews.org/~MichaelDavidCrawford
(Score: 2) by NCommander on Tuesday June 17 2014, @12:32AM
Sure, though please realize that dev resources are somewhat limited, so I have no idea when this would come around (unless of course you'd like to code yourself :-))
Still always moving
(Score: 2) by c0lo on Tuesday June 17 2014, @04:16AM
I... sort-ish... might if you're OK with either PHP or Java riding on top of the database
(the sort-ish: would be a pet project for me, as such, if I'm lucky I'll finish it in one larval stage [catb.org] sitting - at least the "read" part - or I might get dragged into the mundane and put it on hold).
https://www.youtube.com/watch?v=aoFiw2jMy-0 https://soylentnews.org/~MichaelDavidCrawford
(Score: 2) by urza9814 on Tuesday June 17 2014, @01:57PM
THIS. This is a great idea.
I can't count how many times I've sat there *wishing* some site or service had an API. I've done quite a bit of PHP "screen-scraping" of sites to get at the data I wanted. Not sure what I'd use it for, but I've got a bit of a home theater and home automation system set up on my Raspberry Pi, and I'm sure I could find some way to incorporate Soylent into that ;) Would gladly pay a couple bucks a month if I got a simple API with clean JSON-encoded results! Maybe I'd do a 'today's headlines' alarm clock again. Although that just needs RSS really...
Hmm...what could I do with a full API like that....?
That would make it REALLY easy to make some nice mobile interfaces though, that would be great!
(Score: 2) by c0lo on Wednesday June 18 2014, @04:41AM
For my case? Say... I have a bunch of mod points that can't spend most of the time (by the time they expire).
How about something (like a fragment of an app) like: show me messages that I could mod? (don't show me +5-es first, that's no point for me to mod them up again. Don't show me messages in the stories that I posted: I can't mod them anyway).
Yes, such an app would require to know my identity.
If this is done using a token, what SN is "selling" me is not a subscription, but the "token to access a functionality". No hosting, no email or some other ancillaries, but something that has an added value to the very service I'd be using, something that can be quantized in terms of costs - count the calls/measure the traffic - so that the operator can make his/her minds in terms of profitability, etc.
Your needs are good enough with the HTML face that SN dev-team is maintaining? Don't pay anything. You want more: write your app (and to access the data you need in a way that you like) or buy an already written app.
Besides, the capability to tinker with the very things one likes should be the wettest dream of a nerd (in this case, of the programming kind).
https://www.youtube.com/watch?v=aoFiw2jMy-0 https://soylentnews.org/~MichaelDavidCrawford
(Score: 3, Insightful) by zafiro17 on Monday June 16 2014, @04:42PM
Sounds neat. But it also sounds suspiciously like Usenet, which exists, and where you can already go to read and comment on tech stories. Start at comp.misc, which is covering many of the same stories being covered here.
The federated model also sounds somewhat like what is being built over at Pipedot. I don't understand it totally because Bryan hasn't totally spilled the beans on his vision, but it involves a network of federated sites, and I can only assume, some sort of transfer or share of messages or stories among them. It's marching forward, whatever it is!
Someone said, sooner or later, every web forum's new inventions bring it a step closer to Usenet. Here we go!
Dad always thought laughter was the best medicine, which I guess is why several of us died of tuberculosis - Jack Handey
(Score: 2) by VLM on Monday June 16 2014, @04:48PM
"Someone said, sooner or later, every web forum's new inventions bring it a step closer to Usenet. Here we go!"
Someday, web forums and social media will finally approach what we had with emacs and gnus back in the early/mid 90s. Someday.
Other than not having a mouse in the mid 80s, most web forms remind me of my early BBS days in terms of features.
(Score: 0) by Anonymous Coward on Monday June 16 2014, @05:16PM
> Sounds neat. But it also sounds suspiciously like Usenet, which exists,
The key difference is that usenet isn't very dynamic. You can't set up your own NNTP server just by clicking a link. So you have to get a client, find a server, get an account, etc. Plus usenet has spam issues. Usenet might make a reasonable medium for a higher level "protocol" that uses crypto signatures to prevent spam (standard usenet moderation won't stand up to a smart attacker nowadays) but then you are looking at a custom usenet client - well you need that for handling moderations too. I see usenet as useful for archival but not so robust for offloading server costs to users in an automatic way.