Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Interesting find of partially archived MSNTV2 service content
#1
For anyone who doesn't know me well, I have a large interest in the obscure. Technology, media, you name it. WebTV falls into this category of obscure technology I've just fallen head over heels with due to the way the technology was designed and how there was this large crazy infrastructure behind it that managed to operate an entire UX to the end user (no really, WebTV's entire functionality all relies on the servers feeding it HTML and appropriate protocol data). I especially have a weird interest with the MSNTV2, an entirely new upgrade in the WebTV/MSN TV line of hardware, mainly because of the fact it runs on Windows CE and it has a lot of potential for all sorts of weird experiments, but also because from what I've seen it feels less like a browser kiosk and more like a computer.. sort of. Here's the kicker though. WebTV as a whole is so niche that there's very little technical information or archived service material from it floating around still, and the WebTV hacker scene from what I've observed was more focused on trying out the next lame script kiddie XSS hack and were so worried about releasing anything outside of small bits of source code for the service pages (nevermind that they didn't archive other assets used on said pages) that they feared if they even dared release info about anything like server-side exploits or protocol information in WebTV Networks sight line that the scene would be "over". The MSNTV2 suffered a worse fate in that so little technical information about it exists that the most people have done in terms of hacking was mangling it to run Linux, which is cool but nothing too special seeing as the box was a Celeron-based machine. Otherwise it seems everyone stopped caring about MSN TV once the MSNTV2 came onto the scene and was left for the old grandpas and grandmas and weirdo computerphobes that happened to know of its existence, and if I heard about it when it was out I probably wouldn't have cared either. The most that's archived of it right now is sourced from one site containing the box's system sounds, select HTML pages ripped from the system that control how the box operates (again yes they did that), and various test scripts and miscellaneous items. That all changes today.

While looking through the system's HTML code that happened to be ripped and archived I came across a few of the URLs MSNTV2 apparently used for services. Some experimenting with the URLs on Wayback later and at some point I struck gold. I was happy to find out some crucial things like the home page shown when the box logged on and various locations used to host services like Mail, Chat, Newsgroups, etc. have been archived under our noses. Note all domains dedicated to serving user services follow the format "sgX.msntv.msn.com", where X is a number from 1-4 (from my research it appears only 4 of these SG domains ever existed), and as expected the URLs to these services only lead to error pages since I assume they require some sort of authentication from the box. As for how much is archived in general, only a handful of things really made it out. There's the home page as mentioned before, and there's also some sections archived like Sports, Entertainment, and Weather. Also some components used to extend the functionality of pages are completely missing still and I believe that's where the Wayback's help ends. Obviously nothing really helpful got archived either like protocol information or services used to carry out login just to get an idea of how that was handled.

Also a fair warning with the SG archives. It appears it picked up the search pages too, and people's search queries got caught up in that. Let's just say some of the queries are.. in a questionable nature, but I believe said archives have no questionable material linked (nothing that looked legit anyway). Nonetheless I'll link the archives below, view at your own discretion:

http://web.archive.org/web/*/msntv.msn.com*
http://web.archive.org/web/*/sg1.msntv.msn.com*
http://web.archive.org/web/*/sg2.msntv.msn.com*
http://web.archive.org/web/*/sg3.msntv.msn.com*
http://web.archive.org/web/*/sg4.msntv.msn.com*
http://web.archive.org/web/*/mail-sg1.msntv.msn.com*
http://web.archive.org/web/*/mail-sg2.msntv.msn.com*
http://web.archive.org/web/*/mail-sg3.msntv.msn.com*
http://web.archive.org/web/*/mail-sg4.msntv.msn.com*
http://web.archive.org/web/*/sync-sg2.tr...v.msn.com* (contains one capture for a "dr.sync" link that returns blank data. Don't know what its purpose is for or how it works)

UPDATE: I also found a site still floating around that links to various newsgroups on the MSNTV2 service, and the links to them originate from SG domains with 3 other pages with the newsgroup links linking to unique SG domains, strengthening the theory that there were only ever 4 SG domains being used. I'll link all 4 different versions here: SG1 SG2 SG3 SG4

For the most part it seems that all 4 SG domains hosted the same content so there probably wasn't anything special going on with that. What I am interested in is the URL schemes used for various part of the newsgroup service. I'll do my best to cover them on here in as much detail as possible. Let's begin:
  • "sgX.msntv.msn.com/apps/discuss/DiscussLobby.aspx" - No idea what this is exactly for, but probably a shortcut to some lobby newsgroup on the network.
  • "sgX.msntv.msn.com/apps/discuss/BrowseGroups.aspx?branchName={NEWSGROUP}" - This was most likely used to list subgroups of a particular newsgroup (e.g., listing newsgroups under "alt.discuss"). "branchName" would be the newsgroup to look up under.
  • "sgX.msntv.msn.com/apps/discuss/ShowGroup.aspx?groupname={NEWSGROUP}" - This to my knowledge would preview a newsgroup and its contents (posts, etc.). Most newsgroups were listed under "alt.discuss", but MSNTV2 has exclusive newsgroups itself too it seems, like "msntv2.users".
  • "sgX.msntv.msn.com/apps/discuss/ShowPosting.aspx?groupname={NEWSGROUP}&article={number}[&focus={element name}][&fid={another number}]" - Judging from the URL this most likely previewed a specific post/comment in a newsgroup. From the two random newsgroup posts I could find from Googling the URL that appeared to have two variations of this, "article" probably referred to a newsgroup thread ID and "fid" possibly to refer to a reply? "focus" I assume was to tell the server to highlight an interface element to guide the user or something since the values of this parameter I could spot were "previousButton" and "nextButton".
As with the other SG service URLs these were poorly archived and the remaining archives either error out or magically don't exist in Wayback's backlog. It's nice to know the structure of one of these services at least.
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)