Path: csiph.com!v102.xanadu-bbs.net!xanadu-bbs.net!nntp.club.cc.cmu.edu!micro-heart-of-gold.mit.edu!bloom-beacon.mit.edu!bloom-beacon.mit.edu!panix!MIX!Blackbird From: billy@MIX.COM Newsgroups: comp.sys.mac.system Subject: Re: Usenet enters the ring! NNTP vs. (Re: POP3 vs. IMAP) Date: Sat, 10 May 2014 20:29:33 +0000 (UTC) Organization: PANIX Public Access Internet and UNIX, NYC Lines: 141 Message-ID: References: NNTP-Posting-Host: panix5.panix.com X-Trace: reader2.panix.com 1399753773 771 166.84.1.5 (10 May 2014 20:29:33 GMT) X-Complaints-To: abuse@panix.com NNTP-Posting-Date: Sat, 10 May 2014 20:29:33 +0000 (UTC) Xref: csiph.com comp.sys.mac.system:59259 Doc O'Leary writes, quoting me: > > Anyone can post to Youtube. > > How? It certainly doesn't let me. Perhaps you'd like to explicitly state what your problem there is? > > But, you're continually ignoring the fact that Usenet was not created > > nor intended to do this. > > I'm not ignoring it. It's irrelevant given the realities of what Usenet > *is*. The proper approach is to make it stronger for how it *is* used, > not moan and cry about how it's not being used right. As I've said, you'll need to take this up with the IETF. Example - http://tools.ietf.org/html/draft-ietf-usefor-useage-01 And a bit of helpful guidance - http://www.rfc-editor.org/rfcfaq.html If you don't resolve this first, propagation of your desired content will suck, big time. Arts can, among other parameters, be rejected by size, and each server can be set to a specific limit independently. And, even if you do succeed in revising the protocol, that's no guarantee anyone will impliment it. "Make it stronger" involves a significant amount of money - the 14TB+ of daily crap isn't going away, and you want to pile yet more - who will be paying for that? The same people paying for the current load, you say? Nope - a good number of them are already ignoring it. Thus, there's the very real possibility your content will wind up on a mere handful of hosts, like Altopia. If it can get there in the first place. Since you appear not enamored with Google (centralized control of data), this cold, hard fact should give you pause... > > > If I were an ISP, I'd rather have one generic CDN protocol on servers > > > that I control rather than having hundreds of content providers > > > > Where do you get hundreds? > > I use my brain to think about the future. Try it. It's pretty cool! You and your brain should take a good, hard look at all the consolidation in vendors, including CDNs, going on right now. Speaking of (not) cool. http://www.bizety.com/2014/04/07/tough-road-ahead-limelight-networks/ > > Again, it all comes down to who's paying for it. In addition to some > > Usenet admin experience, you'd benefit from a bit of ISP experience as > > well. > > If someone wants to pay me to solve their problems, great. That's kinda > what I do. If they instead just want to complain about their problem > and expect free sympathy . . . get an intern, maybe? Nobody's going to pay you to create yet more problems. And - hard as it is to believe - I'm talking about who pays for the additional resources you want to eat. How could you possibly ignore that, to say nothing of confusing it with what you might get paid? I did a bit of looking around trying to answer my own question, and found, for example, this illuminating bit of info - http://blade.nagaokaut.ac.jp/cgi-bin/scat.rb/ruby/ruby-talk/26555 | Subject: [ruby-talk:26555] Re: Selector Namespaces: A Standard Feature | for Smalltalk? | From: "David Simmons" | Date: Tue, 27 Nov 2001 06:06:56 +0900 | | Doc O'Leary, | | I've read your post/response and I don't know how to respond. You seem to | think that "versioning" is the solution to all the ills when I've pointed | out issues that versioning simply doesn't address. And my posts were merely | augmenting/clarifying items relating the JavaScript 2.0 references cited in | the original post on selector-namespaces. | | I've also made it abundantly clear that versioning is already a supported | language option and that the PRAGMATIC issue is that in REAL situations of | just-in-time deployment it NOT always an option. Why are you ignoring the | realities of software deployment and telling me what I already know about | the ideals of versioning. | | You also prattle on about namespaces with non-specific "dogma", yet you | could not have had experience with "selector namespaces" which ARE NOT just | "namespaces" as we see in other systems. They are a NEW Unifying Feature for | Dynamic Languages that provide a more general solution to what SOME static | languages have provided in the form of "public", "private", "protected". | | They are not meant to be a SUBSTITUTE or REPLACEMENT for versioning [rather | they augment it], they are designed to address REAL issues [which one can | demonstrate "versioning" does not adequately address; as per one (or more) | of my posts] in the area of effective/reliable deployment in 3rd party late | integration scenarios. | | -- Dave S. [www.smallscript.org] Gosh. At this point I'm feeling quite like Mr Simmons did 13+ years ago. > > > Usenet can benefit *everyone*, not just the big players like Google. > > > > _Something_ may benefit everyone. Usenet is not it. You're putting > > the cart before the horse. It'd be best to first find a horse. > > The horse was there long before the cart The fact that the amount of abuse has grown to be in the vicinity of the load you'd like to dump on Usenet does not in any way whatsoever justify it, nor does it render it anything other the abuse it actually is, period. To be 100% clear, "it" immediately above includes your intended traffic. If you don't want to be an abuser, you must first get the protocol revised to include your intended use of it. > but the ISPs killed it. Why should they devote any of their time, money, and resources to handling daily terabyte loads of abusive, inappropiate content? Yea, some of them filter out most of the junk, but others, lacking the modest expertise required to even do that, just throw out the baby with the bath water. > As you well know, even without them, the traffic keeps going up . . . And this justifies nothing, at all. Billy Y.. -- sub #'9+1 ,r0 ; convert ascii byte add #9.+1 ,r0 ; to an integer bcc 20$ ; not a number