Longhorn (heart) RSS
- Posted: Jun 24, 2005 at 10:26AM
- 147 comments
Today Microsoft announced the addition of several new RSS features in the next version of Windows, code-named Longhorn. The Longhorn Browsing and RSS team (the one we interviewed here) is also are announcing a new RSS extension, to be released into Creative
Commons, that lets you do lists in subscriptions.
Check it out, first video demos of Longhorn and IE 7.
There are three demos.
Demo One, at about 23:19. RSS in IE 7 and synchronization with other aggregators (like RSS Bandit)
Demo Two, at about 34:00. Enclosures, calendar integration.
Demo Three, at about 49:50. Amazon integration.
There isn't a download available for this yet. Sorry about that.
Check it out, first video demos of Longhorn and IE 7.
There are three demos.
Demo One, at about 23:19. RSS in IE 7 and synchronization with other aggregators (like RSS Bandit)
Demo Two, at about 34:00. Enclosures, calendar integration.
Demo Three, at about 49:50. Amazon integration.
There isn't a download available for this yet. Sorry about that.
Comments closed
Comments have been closed since this content was published more than 30 days ago, but if you'd like to continue the conversation, please create a new thread in our Forums, or Contact Us and let us know.
More entries in this blog
-
PDC05 Road Trip - Episode 1
-
Longhorn (heart) RSS
-
PDC05 Road Trip - Episode 2
Follow the discussion
Oops, something didn't work.
What does this mean?
Subscriptions allow us send you email notifications when new content is added. You need to be signed in to Channel 9 to use this feature.What does this mean?
Subscriptions allow us send you email notifications when new content is added.sign up for email notifications?
I know that I have a huge bias here, but I really think that MS is doing a really great thing when it comes to RSS. The beautiful part of it is that amount of work that an application developer no longer has to do. I think that RSS is going to be so much broader because MS is putting the work they are doing into the platform. The nice part is that this is not novel work… soon you’ll see these kind of platform investments in every platform. The surprising part is that any one else could have done the work first, but MS did. I think this may be the first time in a long time that MS has done something big that other people will emulate. Because knowing what RSS is and parsing XML for a dev is absolutely useless. It’s like knowing how to handle a TCP/IP packet. It’s the start of a brand new world. And when you can synch your databases, web directories, book marks, photos, calendars, reports, contacts, sales pipelines and everything else you can think of over RSS, you can have announcements like this to thank for kicking it off.
http://ironyuppie.com
Lets hope everyone waits a day or two before someone submits this to /. and Channel 9 goes down again.
Yeah, SlashFUD always posts old news anyhow... I figure it will take them a week to find this.
What's this I hear?
Haven't finished watching the video yet..
mVPstar
http://www.digital-web.com/news/2005/06/microsoft_to_take_rss_five_steps_backwards/
I loved the Amazon demo as well.
I crave more info!
Hmm, RSS Everywhere? I thought RSS was very specfic XML for news updates. Wouldn't XML Everywhere be more appropriate?
I gotta say, great application Walter!
And the whole thing is stupendous and great. This makes my day. The team has a tiger by the tail. The best of luck to y'all in delivering a great set of features for us developers.
♥
RSS
Um, cause it does change?
What, the NT kernel? You've got to be joking...
Oh and Neopets + RSS = uber 1337
Kernel changes in XP
I want to be the first to ask for a Longhorn (heart) RSS t-shirt.
Thanks
Ever heard of Itanium?
You really should do more research on some topics before commenting on them.
Take a look at http://msdn.microsoft.com/longhorn/understanding/rss/default.aspx
Thanks
-Dave
Tiger by the tail? Tiger was released almost two months ago and they've had RSS in Safari that whole time. When is Longhorn gonna be released?
Now all I have to do is to teach my IE friends who have no clue about RSS how to actually use it in order to get their news, including my rants on videoblogs.
Raymond M. Kristiansen
Norway
We should be fine. I found the problem. Let's see if I fixed it...
C
http://www.microsoft.com/presspass/press/2005/jun05/06-24RSSIntegrationPR.mspx
Also, here are some close screenshots of IE7 from Bink.nu:
http://bink.nu/Article4349.bink
-j
"What happens if there are multiple feeds on a page, but they are for different things? For example, lets say my company site has a "News" feed and a "CEO" feed, will IE7 only see the first one listed on the page?"
I am wondering about this too. Some blogs have an RSS feed per category. I could see a simple solution to this. Make that RSS on IE7 be a button with a drop down. Get the first (or default) feed when you click the button. Press the drop down arrow and see all the other feeds you can subscribe to.
Not everyone, but it sure looks like it at the intro to the video. The PM team is often out in front as the "public face" of a product team.
Seriously, the RSS PM team worked super hard on the RSS demos and the Gnomedex event. My dev team was certainly doing our part making the platform solid (IE7 and the RSS bits are real code checked into a real branch of the Longhorn sources) but the Outlook demos, List demos, etc. were all coded by the PMs.
What you see Amar showing is a testament to the strength of our PM team. Big time kudos to them!
Maybe next time I'll be able to be there for Scoble's camera. I'll make sure to bring along some developers.
Basically, our current thinking is that the dropdown is too complex, too much choice. We want the RSS button to be dead simple "boom, RSS preview", not a dropdown to make a secondary selection. So the RSS feed button connects to the first autodiscovered feed on the page. With that implementation, we hope to encourage websites to list their primary feed via autodiscovery rather than a long list.
Note that with many websites, the autodiscovery list is not really multiple contents feeds so much as multiple content formats. That's not really helpful to very many users.
In the Tools / Feeds menu, you can find a cascade of all autodiscovered feeds on the page.
Likewise, some feeds are offered in multiple formats - there might be HTML and plaintext versions of the same feed. Firefox's built-in RSS subscription method displays a menu of all available <link rel> feeds.
One solution for this is to continue to use orange XML buttons (yuck) but it would be nice if IE could offer a simple UI for this.
On other notes... certainly XML is a good way to distribute content. Imagine an RSS patch-distribution system, or virus/spyware-definition-distribution system. But I have to wonder how far generalization-of-RSS should be pushed. I don't agree with the "RSS for everything" mantra. Some things deserve to be their own XML specialization. The location of one's orange sneakers, for example, is probably sufficiently different from news stories to have its own specification. Similarly, a list of active job opportunities might benefit from having it's own XML specification, and might not be sufficiently RSS-ish to work as an RSS generalization. If the RSS spec + extensions becomes too overly huge, it puts a correspondingly huge burden on applications that want to call themselves "RSS clients."
But surely control should be given to the site publisher to decide how best to serve their user profiles. Sites that are less technical in nature could simply scale back to a single feed option. Ian Hickson could continue to offer plaintext and HTML formats of his feed.
EDIT: I don't mean to come off negative. Integrated RSS is very cool.
What control?
Nothing we're doing prevents a publisher from listing multiple feeds via autodiscovery.
IE's RSS button will show the first one in the list because we chose simplicity over flexibility. We had lots of debates about this, and I think we're making the right choice.
Thus he RSS button takes you to the feed, not dropping a menu of everything possible from the publisher. Our current plan is that IE7 will show the full list off a cascading menu item.
... yes it is. I myself have a website on which I was going to offer multiple autodiscovery feeds (linked from the same HTML page, with different content.) IE7's demoed behavior is causing me to change my mind - I'll have to stick to orange XML buttons so that users won't get confused by the RSS button in the IE chrome. Maybe I'll hack up a conditional comment to hide the autodiscovery feeds from IE7... ah, the irony...
Things should be as simple as possible - but no simpler.
-- Albert Einstein
OK the RSS button is on the toolbar and not in the address bar, but the rest is just the same. I was waiting to see a duplication of the content-length slider in Safari, expecially when he started talking about how long the RSS preview page was.
Whats the benifit of Calender items in RSS over vCal? iCal and Sunbird etc already support vCal subscriptions for much the same effect. These standards already exist and are widely used outside of Microsoft's applications, why invent a new one?
Whats the interface that alerts the use when new items are available in the feeds, or does the user have to keep visiting the RSS preview page?
If any application can add items to the users feed list through the API then I expect to see malware automatically subscribing people to feeds that pop up advertising/pr0n every few minutes. What controls are in place to prevent unwanted 3rd parties from adding to the list without permission, a registered RSS application whitelist? LUA wouldn't apply here since it is the user's own data. I certainly hope is isn't accessible to Javascript. I like the idea of a common platform, but you end up with monoculture issues pretty quickly.
IE7 is being released before Longhorn isn't it? So we should see these features sooner than 18 months.
Will there be at least a basic calender componant in Outlook Express for Longhorn, or do you have to have full Outlook?
So when do we get Link Rel tags on Channel9 to subscribe to feeds of individual threads? Firefox already goes looking for them and comes up empty.
--damn language filter.
Why, I made it up, of course... just like I make up all my facts.
Seriously folks, it seems to be a folk attribution. Here's a subtly different version cited on the VS2005 Team System blog
"Einstein reputedly said (although no one is quite sure), that a theory should be as simple as possible, but no simpler."
EDIT: And another version on a mug sold by the Exploratorium in San Francisco:
"Everything should be made as simple as possible, but no simpler."
EDIT2: Einstein usually spoke German, so maybe it got lost in translation. His last words are lost, because his nurse didn't speak German.
So quite frequently a dropdown approach would show either multiple formats (low value to end users) or it would be a single item list (no value to end users). For pages like that, I think our RSS button has the right implementation.
For the less frequent "index of feeds" pages, then your approach of showing multiple feeds on the page is fine, IMHO. If there is no single "best feed" for the whole page, then it just might be a good idea to not use autodiscovery. Your choice.
Agreed. It's a tricky UI problem.
Couldn't you have a button and a drop-down, show the button for sites with one feed, and show the drop-down for sites with multiple feeds?
Or just have a button, and if the site has multiple feeds, have the button pop up a dialog with a list of checkboxes, with only the first one checked by default?
I'm a little confused as to why you think multiple formats will be of low value to end users. Surely that depends heavily on the audience. Suppose the feed is a video feed - I imagine there's value in allowing at least three feed formats (Windows Media, QuickTime, Real...)
Or if the feed is a news site, there's value in allowing auto-sense subscription to any or all of the various "channel" content from the home page.
(Note the media player is an example of exclusive feed options, whereas the news channel is an example of inclusive feed options.)
With IE7 supporting feed autosense, I'm sure the RSS providers will be quick to reassess how their feeds are presented to browser users. Look how quickly rel="alternate" links were added for Firefox, with a mere < 10% browser share.
EDIT: Or perhaps the answer is to allow publishers to specify an OPML file as an alternate link...
A big part of being simple is being predictable. A button that sometimes navigates to the feed and sometimes needs a secondary choice isn't predictable.
We spun around a lot with discussions like these. I'm happy to listen to more ideas on how to make the RSS button be simple yet keep the functionality people want.
By multiple formats, I mean "RSS .92, RSS 1.0, RSS 2.0, ATOM" - a list of transport formats is of low value to the user. I don't mean video vs. text.
Or an option that can change the behaviour from either a simple button or a dropdown box.
Fair enough... so you agree there would be value in providing multiple feeds for different formatting options of the deliverable content (as opposed to merely different formatting options for the markup?)
Perhaps a solution to the RSS vs. ATOM question is for providers to mark up their rel="alternate" links as:
type="application/rss+xml; rssversion=0.92" (text/html; charset=utf8)
type="application/rss+xml; rssversion=1.0"
type="application/rss+xml; rssversion=2.0"
and whatever ATOM is
Then IE could select the most recent version it understands (hopefully when RSS 3 comes out, IE wouldn't try to interpret it until it was patched to have RSS 3 support)
Well its just about which side of the button you click, either on the icon or on the down arrow. People have coped with it in browsers for years, no need to oversimplify it down to just a single button. Adding a user option to change the button appearance is just unneed complexity.
Speaking of which... where are they???
As for an option, "NFW" is what I generally say to anyone that wants to add one more option. If there was any doubt, take a look at our Internet Options dialog - especially the "maze of twisty passages, all alike" of Advanced Options.
Seems better than having to go look in a menu instead.
Thats a Longhorn style window, so the back-forward buttons are no longer part of the toolbar and are now more like part of the window chrome.
In Longhorn Shell Windows + IE7 + WinFX Navigation Applications will share a common style of back-forward button in to top left corner of the chrome.
http://blogs.msdn.com/ie/archive/2005/06/09/427410.aspx
They are in fact putting IE in a very tight straightjacket using NT permissions.
The problem with virtualisation is people still want to move content from IE to their computer. Want to right-click on an image and save it/set it as your desktop wallpaper? That image now has to traverse to the real PC in some manner.
With the locked-down IE you need some method of elevating the IE processes permissions so it temporarily has the rights to write just that particular file to the users profile.
If you are using virtualisation you need a way to alert the host environment of a file transfer request initiated from the virtualised environment.
"Do a good job naming your elements prefixes. No one wants to deal with an element of <wqe412> when <color> is a lot more understandable.
Do a good job with your namespace prefix too. woot is crap when shirt makes more sense (combining this with the above maxim, you've got <shirt:color> instead of <woot:wqe412>)."
Quoted material gratuitously borrowed from http://www.disobey.com/detergent/2002/extendingrss2.
How is it easy? Any action that can be initiated by the IE guest process can also be initiated by exploit code that has taken control of the IE guest process.
How can the host environment detect if the guest IE process has been exploited?
If the host environment accepts file save requests from the guest environment then any exploit code would be able to save files to the real hard disc, which defeats the point of the virtualisation.
I guess it depends on what sort of parser you are using.
If I do a doc.SelectSingleNode("listinfo","http://www.microsoft.com/schemas/rss/core/2005")
its not going to have much luck finding a <a:b> tag, but it will find a <foo:listinfo> tag so long as the namespace reference maps foo to the URI.
There always being a way and there always being a SECURE way are two very different things.
You can't claim to ask a security problem, and then get all worked up when you're given a security answer.
Here's the reality: IE will not only have the low-risk settings, it'll also be using the (oh god, I'm not even going to try and enter the acronym... suffice to say that all content that ISN'T from the machine gets effectively sandboxed using this API set, which was introduced in XP SP2), so not only will the browser be locked down, but ALL content that isn't from the machine'll be kept under lock and key as well.
That answer your question (kinda sorta)?
I'll need to go back and listen to the demo again (gnomedex one), because the second presenter actually covered this.
1. You right click save as in VirtualIE.
2. IE is modified to use a special function
SaveRealFile(filename,sourcesite,pointertofilecontents)
3. The virtual environment traps calls to that function.
4. It causes the standard save dialog to appear on the host OS.
"You are attempting to save a file from the Internet. Do you want to save {filename} from {sourcesite}"
5. The user chooses a location to save that file and the data is streamed out of the virtual environment and onto the host OS disk.
To the end user this is just as seamless as the present method.
However if the virtual browser is compromised, the exploit code could call SaveRealFile("onecoolapp.exe","trustedsitename",pointertoviruscode)
The user is then presented with an unexpected save dialog and may make a bad decision to save the file to the host OS and execute it.
Imagine this scenario.
To save a file to your real profile you would need to.
1. In virtual IE right click and choose save as..
2. Save the file to the disk of the virtual OS..
3. Open an explorer window on the host OS
4. Navigate to \\VirtualIE\SavedFiles\
5. Select and copy the file from there and paste it into "My Documents" on the host OS.
How many people would be able to handle that?
MSN desktop search is more of a limited subset of Longhorn's search features.
Ya i want a longhorn loves rss t-shirt too
Btw i'm keeping track of all the articles and pictures currently that i find on
my blog as well in
this article incase anyone missed some
Regards,
Shital.
http://www.ShitalShah.com
being truly excited about developer possibilities with RSS integration
iCal on OSX has had a feature for years that lets you share your calender, or subscribe to someone elses. You can choose from thousands of business and social groups.
Maybe you just want to know that your chess club meeting next week has be moved to a different time.
You don't have to include all your details in your shared calender, only those that you want other people to know about.
Lots of people seem to like and use this feature,
http://www.apple.com/macosx/features/ical/
Haha...I like people that say "Hey MS, don't copy, INNOVATE!. Look at Apple, they don't copy, it's just you who copies."
8-)
Suggest us, please, some "innovations" you'd like to see coming from MS!
Maybe for you MS products are not exciting but for me, from the developer's point of view, they are...
Have you considered Whidbey, Indigo, Avalon? This is what I call innovation.
OK, now I understand your opinions a bit better
EDIT: What is currently your "wow" feature?
Thats my major WOW feature for Longhorn: "WOW - I get all of those cool things in one box. And they will work hand in hand."
Despite all that as a user I really don't care who came up with what ideas first. If it works and does what I need, I use it. And it is most likely that the next OS by MS codenamed Longhorn will do what I need pretty well and most importantly - better than others.
Take this RSS idea, for example. I bet you a few billion $$ that it has way more enhancements than what we saw on the video presentation today. The only thing we can do now is laugh at the small guys and be anxious for these API's to be released.
My arms hurt.
Steve.
Exposed through managed code?
It's exposed through a documented set of API's. It shouldn't matter if it's managed or not.
Yes, it does matter. Want to know why? Because we want to see CODE!
I have to disagree. If that API is managed or not is crucial. And it wouldn't really make sense to make hat API an umanaged API. BTW: some of the demos shown in that video are said to be written in c# ....
I use FF and I have to admit that it's far better than current IE 6 but I'm also excited about IE 7; it won't get just RSS and tabs, there's gonna be much more in it (I hope
This is more than IE 7. If you think its just IE 7 you didnt watch the whole video. they are integrating RSS into the entire platform. It will be easier for developers and content creators to integrate RSS everywhere. I have heard that RSS will destroy traditional web browsing. I doubt it, it will be complimentary people have to search and find the content before they subscribe. Im a Netscape 8 user but I cant wait to see IE 7. I think Microsoft is doing some very interesting and innovative things with RSS including IE 7. I cant wait to see if IE7 RTM does indeed pass the Acid2 test. Im not expecting the betas to, but maybe we will be surprised. I dont like the new team name, "Longhorn Browsing and RSS", They should call it the "Windows Browsing and RSS" team
it's not that MSFT just got hip to RSS.
thats about the last thing they were telling you.
1) they will ship a set of classes and namespaces to make programming RSS simpler for windows. The goal is to make the basic plumbing "just work" for you and then you do not have to keep doing that part.
2) they see that enclosures are a good thing that can be used much more creativly than they are used today.
3) they are publishing a set of guidelines for how they think RSS can work with more enclosure types.
4) the windows classes will handle a whole slew of enclosures "out of the box" for you.
5) the publising of this work follows creative commons cause they want to "Play nice" with the rest of the RSS world.
ok, so they did not show us how to do warp speed or time travel but the things I just listed see very good to me....
I do understand that you feel the way you do. Allthough MS has been very open about the bowels of Longorn there is not yet too much thrilling to see from a plain user perspective. This antagonism paired with MJFism can lead to prejudices or to the kind of feeling that you have. One needs to keep in mind that longhorn is not even beta yet ( I know, it makes me sigh too ...)
Form the developer's perspective however there is already quite long list of promising and innovative technolgies and concepts publically known and ready to explore. Further as a developer I find it rather easy to imagine what could be implemented with them.
You read that MS ? That's what happens when you start milking the longhorn while it is still a calf.
All I know is, Longhorn users will have a lot of add on downloading to do between 2006 and 2008.
I'm all for playing well, but you have to simplify when transitioning it to the mass market.
Nobody would have to pay me to wear one of those shirts. They just look cool. How much would I need to pay to get one?
Sign me up as well. =)
I think there's a difference, from a security standpoint, between calling code via interop and calling managed API's to the .NET Framework. I *KNOW* there's a big difference between the two methods from the PITA standpoint.
I agree 100%, I would love to have one of those shirts, I asked for one earlier.
The thats why they are building RSS into Longhorn at platform level so its exposed to every developer building applications on Longhorn. And I know that some of Groove Networks technology is going to tap into this for P2p scenarios.
P.S. That comment on The Centre de Pompidou was unforgivable, I've queued for two hours to get in and its a magnificent building. I'm glad that guy isn't on the Avalon team.
$mplayer -dumpstream mms://wm.microsoft.com/ms/msnse/0506/25055/RSS/rss_longhorn_platform_2005_MBR.wmv
RSS is too specific. Firefox made agood choice when they changed the RSS icon into something none format specific. IE7 should do something like that. What if there was a format that ended up superior and simpler than RSS and Microsoft liked it? The RSS button would seem obsolete then.
Syndicate or feed or something better and future proof should be used.
What you have to think about is the average home user in this situation. If they see that their OS (which is most likely going to be Windows) has a web browser built in (IE), they are probably going to use that instead of researching other broswers and spending time testing them all. As IE hasn't had features like RSS and tabs before, the average home user won't have had the chance to experience them. Now that Microsoft has taken the steps to change that, they are opening up a new world to that section of the public. It might not be revolutionary, but you have to say that at least they are willing to show average users modern advances.
However, from a developer's point of view, I totally agree with you. We access the information that shows us that there are browsers that are better. Yes, we are more concerned about standards compliance while the average user isn't. We know how important this can be to successfully render pages. This is one of the main things I would also like to see from new versions of IE. Standards are there for a reason.
What I'm saying is that IE do not have to innovate to improve their browser. Who says you have to innovate? Sure, I would like to see it, but it's not essential when creating a good browser.
This can/will have very far reaching impacts.....will be fun to watch it evolve out of the "blog" space and into the day to day lives of everyone...
Try this.
Avalanche is a research paper, nothing more. Drop the conspiracy theory and back away slowly, and no one will get hurt.
You know...I respect the guy for the great work he's done with BT, but he comes off as being quite rude and arrogant.
"First of all, I'd like to clarify that Avalanche is vaporware."
Vaporware? It's an MSR research project, not a product that's going to ship anytime soon.
"As you've probably figured out by now, I think that paper is complete garbage."
Again, this statement was completely uncalled for. Research communities often involve groups critiquing each other's work, and everything in that post before this line was all well and good. Calling it "garbage" goes a little over the top though. Constructive criticism is good, but that part of the post was just childish.
Talk about a load of rubbish.
You have obviously not re-checked linux for a number of years.
Novell themselves have moved to a pure SUSE desktop for all employees.
Maybe not all aspects are covered, say like GAMES (which MS OSes are only good for).
--
Lets get one thing straight. MS Dont care about their browser.. they never have.
They care only about one thing. The Monopoly. ie The market share..
They caved in the IE dev group once they won the race against Netscape.. and they only won that by embedding IE into their OS so badly you couldn't remove it and every lame user who wanted a PC got Winblows with it and never thought to think outside the square to check up on other browsers.. which there in and of itself is a bad situation to be in.
To top it off the only time the IE dev team was put back into action was when their market share was being threatend by 2 things. 1: Hackers, 2: Firefox.
RSS Feeds being included in IE7.. Big fracken deal IMHO.. M$ are behind the 8 ball yet again and they're making out like they are the first to think of it.
SORRY.. but DONT BELIEVE THE HYPE.. M$ Is all about marketing and HYPE and brainwashing people to think they're on to it.
Anyone who thinks MS has been good for the industry is an idiot in my books.. complete and utter idiot.
M$ Called AMIGA a games PC... Why? because it had sound and graphics to a level that made games great but at it's core it did those well because hardware and software talked closely.
Today we have M$ PCs all reverting to a AMIGA situation (well as close as one can get when you're multi threading and not multi tasking). They have GFX cards of emmense quality, they have SND cards of greatness.. and yet this is considered normal now.. but in the amiga days.. noooooo that was frowned on - you didnt need sound or high qual gfx to write a word document did you..
What ever they can twist to their will MS will do so. Including your meagre pathetic brains that wish to bow in subserivance and lack of lateral thinking or investigation on prior art to ideas. Nono.. its ok.. you keep thinking MS know their sh*t and believe it without question.
As for development environments after sampling VB and its inability to make an executable without the requirements for DLLs - that put me off M$ development environments for life. But thats ok.. the dumber you make the developer the easier it is to get developers.
[lifetime rant ongoing - sick of MS lies upon lies upon lies.. 20 years of it makes you sick to death.. anyone ever heard about the story of the boy who cried wolf??? well M$ have cried it toooo many times so now I'm crying only for the death of M$]
And kids... remember one thing.. MS OSes ARE ONLY GOOD FOR GAMEs (*cough* and being hacked *cough*)
Yeah, I dont think there's any online community that whines as much as Slashdot.
Not sure how you would make it work...but for example, let's say I were at Gnomedex and subscribed to the schedule. I'd want to see a transparent version of the Gnomedex schedule on top of my personal schedule (the base)...instead of side-by-side.
Probably not an important feature, but it would have a lot of coolness factor.
Anyway, RSS support in the OS...exciting stuff.
Oh and Blogger hats? Isnt Blogger Google?
Multiple major original technologies get dropped from Longhorn (WinFS, Avalon, Indigo, Monad, even the whole .NET2 CLR) and Scoble say "it's OK, there's still great stuff in there, we just can't tell you about it yet". Then this comes out - Longhorn <hearts> RSS. A trivial (sorry, really simple) technology that essentially belongs in userspace and is already well served on existing versions of Windows and other OSes by many 3rd party packages and utilities.
This story should have disappeared without trace and I note that it more or less did in the mainstream media (even the mainstream IT media), but online and particularly among bloggers it runs and runs.
Mainstream if you count the tech press, which makes up the majority of those news items, as mainstream.
I'm not sure what that's supposed to prove except that online news sites are susceptible to press-release journalism. Even BBC News online which usually gives big coverage to anything involving Microsoft didn't make it a "picture story" on the Technology page.
But that's besides the point. My contention is that RSS is a "me too" technology. It's nice to have - maybe it's even important to have, but nobody is going to upgrade to Longhorn becuase it has some RSS support. If this is the big news about stuff that is making it into Longhorn then Microsoft is in big trouble.
All those technologies are developer APIs. And WinFX (Avalon, Indigo, and .NET 2.0) are certainly in there. I don't think anyone outside Microsoft knows what's going in Longhorn in terms of user features. You've heard them go on about a couple features in the shell like desktop search but Longhorn is about big sweeping changes and polish not tons of individual features (although there are many).
There are hundreds of teams inside Longhorn all working on new features or improving existing Windows components. You think when Longhorn comes out next year, all those developers were sitting idle for the past 5 years?
What ever happened to this idea that strongly typed datasets over web services would allow people to consume and use schemas (like WinFS). RSS is not a low level thing, nor is it object oriented. You call a URL and get a calendar back but what about paging, sorting, data integrity (strong typing with schemas[int,string,bool,etc.]), and security.
What about posting data? Our calendars are always read only? Our friends list is always read only? Or do we just go ahead and send an untyped RSS data structure back to some aspx page? This is the slop that I can't stand.
Peer to peer is great but peer to peer can also incorporate schemas to share not only the binary data but the types that are transported between the consumers and senders. There is so much potential out there with WinFS. Lets not get all side tracked with RSS and the fact that the community of non object oriented developers enjoy consuming "items" because they're so "easy". I think RSS is a great add on but to me, it's just another XML structure that allows people, that don't understand what strongly typed objects give you, to create user interfaces that just display data. Anyone can create a wrapper for windows media player
Another problem is the fact that outlook, Active Directory, Exchange server, WINS, DNS, and many other products run on old Microsoft Access databases. The APIs to these things are incomplete, antiquated, and very difficult to use. You also get data corruption from these things. Once WinFS is out, we'll see these critical windows services moving into SQL Server 2005 express or some ambiguous flavor that comes with the install of the application. Then we'll need a way to expose the schemas for these services and applications so that we can consume, send data, and communicate with the APIs.
I will be much more impressed when WinFS releases their library of schemas that work with all of the windows applications and later other languages start adopting XSD. Then you RSS monkeys can go and create all of your own classes in whatever language you want and I'll be over here autogenerating my classes and DataSets with prebuilt schemas. That's when we're off to the races folks!
T
C
Thank you
(Quote from video) "RSS came from blogs"
??
What? RSS was born from Netscape in an effort to broadcast news to it's users.
While I applaud the team adding support directly to Loghorn, no matter who you are, you should give credit where credit is due, even if it was your competitor (who later R.I.P'ed)
Some suggestions as to who would want to publish their calendar. Getting any of these automatically into a calendar would be obvious steps:
- NYSE, London/Tokyo Stock Exchange, etc, showing market holidays. From experience, it is a total pain having to enter by hand a list of holidays for stock exchanges in multiple countries (e.g. http://www.nyse.com/Frameset.html?displayPage=/about/1022963613686.html for NYSE holdays, not yet RSS);
- Governments listing pblic holidays (e.g. http://www.opm.gov/fedhol/2005.asp for US federal holidays, not yet RSS);
- Corporates listing their upcoming investor relations announcements. Investors and the media subscribe to companies' email lists, to get announcements. Some companies even RSS-enable their media releases. Taking it one step further, by putting the calendar into RSS would be the ultimate (e.g. http://www.vedior.com/investor-relations/financial-agenda.asp);
- This would be a huge project, but you could even have share registries and stock exchanges putting key dates on RSS: e.g. ex-dividend dates, dividend pay dates, share tender dates, deadlines for AGM voting, and so on for all their companies.
So, both you and the team are correct.
How does this work if one have several machines?
I use NewsGator, because it aggregates to Outlook/Exchange, and if I read a RSS post on one machine then it is automatically marked as read on other machines.
With 30 RSS Feeds, one doesn't want to read the same think twice.
I'm not using a roaming profile, because my home PC is not part of my company’s domain.
If this is not supported, I think I'll stick with NewsGator.
Thomas
PS: 5 Program Manager for this job? Why is this so hard? Maybe they should learn about this think called .NET.
Orbit, the ‘other operating systems vs msft’ case has already happened, you know, back in the late 90’s.
see-
http://radio.weblogs.com/0001011/2005/06/29.html#a10502
However I think it is important that the common store can easily be kept in sync with multiple machines. In fact the main reason I moved to Small Business Server for my home network was to keep outlook in sync between my main PC and Tablet.
However I feel that a sync feature for the RSS common store should be a built in from the beginning. Including the ability to sync your office and home machine, like in RSS Bandit, as well as a more continuous sync option, like you get with Outlook and Exchange.
First off, I definitely think podcasting is a fad. Unlike blogs, where you can get data from a thousand people in a day and make sense of it, podcasting requires real time to take in, and has far lower information density than a blog.
There's a temporary 'cool' factor to hearing the voices behind the words, or putting your words out there. However I believe that many people will only put up with shoddy production values in a text-based blog and now, in the early days of 'podcasts'. Eventually the 'marketplace of ears' will gravitate towards professionally qualified speakers, who will (and already do, from what I can see) dominate the podwaves. There's already an 'abandonment' issue with blogs, and I think that the abandonment rate for podcasts will be far, far higher.
Also, podcasting will never be the 'everyman' movement that blogs have the capability to be, and we shouldn't want it to be. There may indeed end up being more podcasts than there are radio stations, but the 'curve' of who gets listened to will be extra-sharp. This makes it a 'fad' in my eyes, at least.
(We shouldn't want podcasts to be the everyman movement, because it is nearly impossible to extract useful information out of them. A recorded talk show that talks about the decisions of the Supreme Court isn't searchable, indexable, hyperlinkable, or anything else that makes the web...well, the WEB.)
Anyway, I think Microsoft is doing some very cool things with their RSS support, despite my normal dislike for them. That said, I hope that Microsoft takes to heart the idea that this 'feed store' that they are creating should be able to be exposable to the network via an open protocol, that anyone can implement client or server for.
This way, I could kick back on my Mac, running NetNewsWire, and have my subscription feed be the same as when I'm sitting on my Windows box at home, or at work, running a Windows aggregator. I can subsribe from any box, and all my boxes know about it. You have the opportunity to fix the 'bookmark' problem now, before it happens all over again.
One extremely important place that this comes in, is in password-protected syndicated content. It's only a small subset now, but as pay-for site-subscription models and even subscribing to sensitive application data (like the log file aggregator mentioned at Gnomedex) become more popular, you really don't want to be putting those feeds someplace where the passwords are exposed.
Putting it all together, I actually think that Microsoft has the opportunity here to define (or work with the community to define) a good, not-Windows-specific feed storage and retrieval protocol, also opened up under a community-friendly license.
So.
Pipe dream?
-- Morgan Schweers
Interesting to see the IE7 integration - and subscribing to 'calendar' and all - just GREAT..
Acronym - there's another RSS that I've been working with - REPORTING SERVICES SCRIPT
Part of our dev project involves Reporting Services (RDL) - and have some script files to deploy - using 'rs.exe' - Report Server Script Host
And these script files are called RSS as well !!
Oh well - just another 'TLA'... ('Three Letter Acronym')
Well,
but are they allowing going multiple steps backward by a kind of drop-down menu. It doesn't interrest me if common users don't uses them - I have a very heavy use of it. Does the current model allow that? I can't see that from these screenshots.
The users who don't understand (I dont say "need" because that's not the truth) this behavior, are currently just ignoring that behavior. So I don't see the need - and discourage - to remove the drop-down menu functionality.
Greg Cobb
B.S. in Computer Engineering at the University of Texas
I started to download the file thinking that it would be PodCast sized and came back to discover that it was currently at 156MB and still going.
Lift your game guys!
In addition to IE7/Vista having RSS its one thing to have a constant supply of information, but quite another when dealing with the archiving and then easy access to those stories. Consider the best way of storing and searching for those news articles, online or locally? I prefer locally, which means saving the story’s picture and text in a Word file, manually.
Most news articles have an attractive point of reference, this being the little bit of eye candy which visually reproduces the topic of discussion. This often is a far easier point of reference to retain in unconscious memory, so when searching manually through pages and pages of old stories in bookmarks or online often a picture is the best source of reference compared to text.
How can an image be a story? Simple, it can be done today, but not with all image types, only those that allow for meta-data fields, luckily most images online are JPEG’s:
Find an online news article that has a JPEG picture, save this with the relevant file name of the article e.g. ‘Worlds Strongest Woman’ now in the properties/meta-data fields copy & paste the articles body text, then add some more info like category, author, date, URL and a title (if so desired , but not a necessity because of the files name)
Now the image contains the story, thus the story is the image, the catch is to make this a one-click process for the user. To do this an option in Vista /IE7 would allow the user to hover over a picture and select ‘Save Image as an Article’, the browser then automatically saves the image with the title of the story automatically adding meta-data such as:
Title - in the title field (same at file name)
Author of the article
Date in D/M/Y format specific to users region (refers to users computer) come on lets make this intelligent or have the date indisputably displayed as 27th July 2005
URL - for future reference
The story itself - naturally
Obviously the whole idea is that the browser (or whatever we will be using) will collect all the relevant data correctly every time with no mistakes, but I think some users would like more control, so (if configured) the browser would:
i) In a separate window, display a fit-to-page view of the article with sections highlighted to denote what will be collated into the picture, much like the print preview in IE but with boxes.
ii(a) The browser could use pointers or tags in the code to automatically denote what text to save into the specific image fields
ii(b) For a ‘dumb’ webpage then the whole article is boxed and highlighted in a segmented fashion with each text box user selectable, to be included or not
ii(c) The computer would learn what kind of stuff I like through my habits so I could eventually just trust it automatically, especially on regular sites where form past operations the pattern is obvious (a batch feature would be nice)
**Maybe in the options tab have a hi/low collation setting so ‘Low’ = not much info auto-selected, and ‘High’ = most of the pages text.**
Most article images online are in JPEG format. However to avoid the problems of images with no meta-data fields, thus scuppering the process and displaying “Sorry this file contains no room for meta-data, no-can-do”, provide the option when hovering over e.g. a GIF or BMP file to convert the file on the fly to a JPEG so the process could begin.
It would also be useful if the process validated weblinks, nothing like coming back to a story in a year’s time only to find the link added at the time was a dud.
A similar feature which would instantly create a Word document from articles that had no pictures, would work again based on the same principal by collating all the relevant information, this would be a god send - maybe insert the websites logo as a picture (but no adverts!!)
Once locally saved on the computer it would be interesting to see how the ‘image stories’ are manipulated by Vista. I would like to see Vista use the meta-data info effectively and behave according to what is in a files properties.
Not simply organising by date/time/author, add value to image files with their property fields filled by labelling them with a superimposed transparent corner strip to signify ‘it’s a story’ which separates it from ‘genuine’ images. Of course as this becomes commonplace the OS would not have to do this as meta-data becomes richer the OS would take it for granted.
1. Simple features like a menu appearing when highlighted - conjuring up options e.g. an option to visit the website or search for similar stories but more recent.
2. Search for articles of the same category by that author etc...
3. Add the stories to a portable device and use text to speech to dictate the story to me while I look at the picture or out the window
Now don’t tell me this wouldn’t be a better way of saving news stories rather than as hyperlinks, which get lost and are hard to navigate.
Samuel, UK
Not much useful knowledge on the blog.And the blog is leisurely to reveal, albeit interesting.
Remove this comment
Remove this thread
Close