Thursday, March 15, 2007
LDAP
The origins of LDAP begin with the International Telecommunication Union (ITU) based in Geneva. ITU began setting email standards which required a directory of names (and other information) that could be accessed across networks in a hierarchical fashion not dissimilar to DNS. The result of their work resulted in the X.500 series of standards which defined DAP (Directory Access Protocol), the protocol for accessing a networked directory service. Tim Howes, Steve Kille and Wengyik Yeong saw a better way to achieve the aims of ITU and published a proposal entitled X.500 Lightweight Directory Access Protocol in RFC 1487 during July 1993. Their abstract read as follows: The protocol described in this document is designed to provide access to the Directory while not incurring the resource requirements of the Directory Access Protocol (DAP). This protocol is specifically targeted at simple management applications and browser applications that provide simple read/write interactive access to the Directory, and is intended to be a complement to the DAP itself. Tim Howes of the University of Michigan led the development of LDAP supported by the National Science Foundation. From 1993 to 1997, the Internet Engineering Task Force (IETF) published 13 RFCs related to LDAP and dozens since. In essence, the University of Michigan invented and created LDAP. Two open source projects resulted from the work at the University of Michigan – OpenLDAP and the Fedora Directory Server. On April 22, 1996, in a press release, the University of Michigan announced that "Netscape (would) incorporate Lightweight Directory Access Protocol (LDAP) directory service technology developed at the U-M into its World-Wide Web software." The University felt they had taken "a giant step toward having (a) global directory service on the Internet." Netscape and the University partnered to create the Netscape Directory Server (NDS), which became the leading LDAP application in commercial use. Interestingly enough, Red Hat purchased NDS from AOL and open sourced its code. That resulted in delivery of Fedora Directory Server (FDS) to the Linux community. The OpenLDAP Project was started in 1998 by Kurt Zeilenga. The project started by cloning the LDAP source code from the University Of Michigan. When attempting to access the original LDAP project, the page redirects you to the OpenLDAP project. In essence, one could speculate that the OpenLDAP project became the successor of the University of Michigan project. Subsequent Internet protocols have their origins in LDAP including the XML Enabled Directory (XED), Directory Services Markup Language (DSML), the Service Location Protocol (SLP) and the Service Provisioning Markup Language (SPML). Of course, this short discussion requires mention of one other subsequent development - Active Directory (AD). Redmond utilized LDAP protocols to create AD, which demonstrates the remarkable potential of the founders' (Tim Howes, Steve Kille and Wengyik Yeong) work, which began back in 1993. If you haven't delved into AD, you owe it to yourself to do so. The Linux community could use the AD blueprint to manage enterprises. That could extend the historical discussion of LDAP even further.
Miscrosoft's crippled OS
Microsoft is applying for a patent for an operating system that starts out crippled. You must pay to do things like take the throttle off network speed, disk access, install drivers, install software, and more. Trust me, I rarely even visit Groklaw, even if this is my second consecutive blog entry with a link to a Groklaw article. My VarLinux.org readers posted the article, A Brave New Modular World, and I had to share it. As the article states, the fact that Microsoft is attempting to patent a cripple technique isn't the real story. It is more likely Microsoft is applying for this patent in order to float the idea and see what kind of reaction it gets. Regardless, if Vista is any indicator, it seems very realistic that the next version of Windows will implement something like this. Vista already shifts into throttled cripple mode if it detects what it believes is pirated video. The idea behind this patent is that the next Windows could start in cripple mode, and the only way to get it to perform normally is to pay Microsoft to unlock "features" that most right-thinking people assume the operating system should provide by default. The article paints a scenario where this could actually discourage people from switching to Linux from Windows. After all, Dr. Stupid opines, if you pay for a 3 year subscription to unlock the throttling and get access to updates, you'd want to get your money's worth and stick with Windows for 3 years. Here's a question for you: Assuming Microsoft follows through with this plan or something like it, do you think this would really lock people in? Or would it drive people to Linux (or just about anything else besides Windows)? (Disclaimer: VarLinux.org is totally non-profit, and I have no association with Groklaw, so the links are not obfuscated plugs.)
Apple and Linux
That's the question that occurs to me as I read this piece in Roughly Drafted. It's about how Apple is kicking Microsoft's butt at the high end of the desktop market, and how Microsoft seems to be bumbling its way out of desktop hegemony anyway. Linux is mentioned only twice in this long piece, but the harbingery of the references are significant. Here's the enclosing quote: Combined with the dominance of the iPod over devices using Microsoft's PlaysForSure, the imminent goring of Windows Mobile by the iPhone, and the shift of support across the industry from Windows to Linux in servers, the days of Microsoft's monopolistic grip on the desktop are winding down. Apple doesn't have to take a majority share of the desktop market to win, it only needs to take the most valuable segments of the market. Once that happens, Microsoft will be forced to choose whether it wants to battle Mac OS X for control of the slick consumer desktop, or repurpose Windows as a cheaper, mass market alternative to Linux in corporate sales. And, at some point, consumer sales as well. Because Apple will never make a cheap desktop. And Microsoft OEMs will at some point break clear of their exclusive partnerships with Redmond. The market will demand nothing less — as Glyn Moody has been pointing out here lately. Thanks to the work of free and independent developers of all sorts, the sum of all device drivers and applications for Linux desktops will inevitably reach a tipping point. Dell or HP or Lenovo or some other company will start making cheap Linux-branded desktops and laptops that are easy to use and well-supported. nVidia and ATI will follow Intel and AMD in the march toward the marketplace. Phones will follow. Because Apple will also never make a cheap and open phone. But they will help open the market for one. Just like they're opening the market for cheap and open desktops by breaking Microsoft's hold on the high end.
Subscribe to:
Posts (Atom)