For an installation even one-tenth of this size, X terminals and big servers are the way to go. And no, the terminals don't need to be fast. We use P133s with 32MB just fine. Administering two, or even ten servers is a whole lot easier than administering 2500 independent workstations, even with the slickest use of NIS and NFS to pull off centralized configuration and management.
How much document sharing will there be with organizations that use other (read: Microsoft or Corel) office software? For interoperability and the shallowest learning curve for MS Office users, Corel WordPerfect Office and StarOffice are the best choices. Both will read existing MS Office files well enough, and will export new files in that format well enough, too. Neither is up to the task of heavy back-and-forth collaboration with MS Office users, but that's true of any office suite on any platform. StarOffice can't deal with most WordPerfect Office files. This would be a non-issue under 98% of circumstances, as so little of the world really uses WordPerfect anymore. However, as a state court system, you're part of the 2%: WordPerfect still has a strong presence at law offices, so interview a representative sample of users and managers to find out whether they currently do receive a notable number of WordPerfect files via e-mail or on disk. Frankly, they probably don't, instead getting them via fax or in hardcopy. So it may well not be an issue.
This answered and all other things being equal, I'd opt for StarOffice; Sun is in better financial shape these days than Corel, and bulky though StarOffice is, it's also natively written for Unix/Linux. It's also got more features that suit it for network and large-environment use. Both Corel and StarOffice are available in identical versions for Windows, so the laptop brigade can work seamlessly withe the terminal crowd.
For another thing, many job functions don't require an office suite at all. Don't assume everyone needs one, and don't just reflexively give one out, even if it's free of license fees like StarOffice. If someone just sends simple faxes and email, that's all they should be able to do. If someone simply accesses an AS/400 or mainframe and works with e-mail, they need access to nothing more than a web browser for e-mail (or perhaps Netscape Communicator with its IMAP mail support), and a tn5250 emulator. For sending faxes, use email-to-fax and fax-to-email gateways. Hylafax is your friend. And incidentally, StarOffice has nice hooks for printing and "emailing" through networked Hylafax servers. It can also sync with Palm gizmos.
For more complex environments, StarOffice has some further advantages. Enterprise-caliber support contracts are available, as are user and administration courses. Macros and scripts for it can be written in Javascript or in VBA. MS Office VBA scripts themselves won't work, but the skills some users and managers may have can be leveraged very easily. For another thing, it can be scripted with and interact with Java. So what? Ah, here's the nifty enterprise-caliber part: not only can StarOffice 5.2 access ODBC databases. It can also access any JDBC data source--which means pretty much any database on the planet, regardless of OS. In addition, you can take advantage of Java toolkits and SDKs for all sorts of things. For example, IBM has a toolkit for Java access for AS/400 client APIs. Want to add a menu item to StarOffice for retrieving data directly from a mainframe into a spreadsheet? Or linking calendar items (have I mentioned StarOffice's Outlook-like group calendaring?) dirtectly to AS/400 screens? You can. In ways that can be reused on other platforms and environments.
What Linux distribution you use is the least important piece of this. Choose something that offers easy creation of kickstart disks, to ease installation on new machines, and that offers good, reliable security upgrades. Me, I'd go with something that offers decent commercial support contracts. The terminals won't need any such nonsense, but it sure is nice to know you can get an engineer on the phone if a $50,000 server has a memory leak you can't squash. There are several ways to do this: you can go with a Linux vendor that offers contracts, like RedHat, or with a hardware vendor that sells Linux OS support on its hardware, like an IBM or VA Linux. For this reason, Mandrake comes out weaker than RedHat. It's not about the quality of the default installer or the number of "extras" on the CD. It's about maintainability going forward and the quality of support you can buy for those times when Usenet and online documentation cost too much downtime.
Depending on what you need in the way of file storage or backend applications, you may well want to look into a commercial Unix (say, Solaris, which runs StarOffice splendidly) on the backend. Linux is great, but if your needs really call for a SAN (and it doesn't sound like they do), or you want to go with one gigantic 24-CPU server instead of several 4-CPU ones, Linux may not cut it. Don't compromise your implementation for politics. Keep in mind that at this level, Linux is Unix is Unix, and that things like data and email and so forth can move fluidly between flavors without a moment's thought. Linux can certainly support 2500 users well; it can support tens of thousands of users well, as most large universities can tell you. There are also things it can do that something like Solaris can't, such as attach seamlessly to Windows file shares. But plan your applications and your network before you make the final OS decision, so the OS doesn't force compromises.
And another nice thing about these Linux X terminals is their flexibility. Just need green-screen VT102/3270/5250 access in the mailroom? Can do. Need to mix in some Windows-only applications after all? Set up a Metaframe server and give the X terminals access to an ICA client. Want users to be able to save to floppy, or attach a barcode reader? You can. With no changes on the terminals themselves.
Corel can't market their way out of a paper bag. They've got ugly ads, ugly packaging and lowball pricing that undermines confidence in the product.
Add to that a product strategy that seems to involve implementing features nobody ever asked for because they have orphaned technologies lying around, while leaving gnawing feature gaps intact year after year.
They're a slowly sinking ship. Their brand is synonymous with rescued orphanware, with shovelware, and with high-end software that only amateurs buy. It's one thing to be a niche vendor. It's quite another when the only niches you dominate are for word processing software for law offices, and graphic designers who have no money.
There's nothing special happening with Linux-related software reviews. Damn near all the reviews in the broad-audience print and web publications are superficial, usually on the positive side.
You need only read something by some of the exceptions like the New York Times's Peter Lewis or Byte's Jerry Pournelle to get a sense of proportion.
In the major computer media from IDG, ZD and CMP, reviews of office suites or $600 graphics tools seldom go beyond a checklist of features and a few comments on interface design. How often does it (StarOffice, MS Office) crash? How often does it (Adobe Illustrator 9) mangle files created with it to the point that they can't be opened?
Large-scale databases and high-end application servers sometimes get a proper review, with a realistic test environment, a significant amount of use, and real-world legacy data. But desktop operating systems and applications--even expensive ones, like design software and development tools--just get a review based on firing it up a couple of times and a look through the feature set.
Sure, that UML tool integrates with PVCS, but does it integrate well in an active team environmet? How's the interface? Does it freeze up when there is network latency? Yes, that desktop database can support tables with millikns of rows, but how did it perform on complex queries? And how did its "multiuser" features fare when you try that same query when three other users are making queries?
Linux-related reviews seem no better or worse. Caldera OpenDesktop installs easily? Great. How well preconfigured was Netscape? Did the PPP dialer setup utility require odd gymnastics like firing up a terminal window and running it as root from a command line? Did you have trouble accessing the update and patch download site? How responsive and knowledgable is the phone and email support staff you're paying for? Were tasks like printing a screenshot or installing new hardware easy or difficult?
If there is a general disconnect between the low polish level of many Linux desktop apps and the high praise Linux gets in the resulting review, that probably has something to do with the remarkable stability of Linux itself compared to the major consumer operating systems. No antialiased fonts? Awkward printer setup? But gee! The darn thing ran for four the weeks of the evaluation without a reboot.
It's tough to get a meaningful review of, say, Photoshop, from someone who is a product reviewer by trade, and not a full-time graphic artist. Even if the reviewer was an artist in a past job, s/he is seldom going to give the application the hard workout they would have in the course of real-world use. Instead, we get charts giving rendering times for specific effects, a database-style test suite applied to a creative tool, which should be more about flexibility, interface design and stability.
Enterprise applications are easier to review, since you can draw meaningful conclusions from a rigid, numerically quantifiable test suite. But even here, reviewers grab onto the superficial to make major pronouncements. So-and-so's web-based server management console is lacking? But what if most users of the application use a command line to manage the product anyway, as with many databases and web and application servers?
Pirhana is part of the standard RH 6.2 distribution. What you're getting here is a year of support; but more key to this, you're also getting consulting services: For your $2000, they will provide, via phone and email, the exact configuration information necessary to set up a node's worth of clustering on your network.
Comfortable with documentation and understand clustering already? Then get the $150 SSL-enhanced version of 6.2 and set it up yourself. Everything you need is there.
Don't need an RSA-licensed Apache-SSL in the node? Then grab Redhat 6.2 for free and configure Pirhana (which is included) at no cost.
Consulting costs money. If you don't need consulting, good for you. High-availability clustering is free.
So there's another company trying to compete with DMX, MusicChoice, Muzak and SkyRadio. Their one hook is that they're trying to be the first to get portable and mobile receivers made, which will go thr way of the dodo, quadraphonics, 8-track and AM stereo if the receivers are designed only to hook to their system.
Satellite and cable narrowcast "radio" stations have been around for some years now. And yes, while DMX and Sky are available to digital cable and home sattelite customers at home, their core revenue source is commercial subscribers: shops, restaurants, offices, buildings and so forth, which pay a higher fee.
There's certainly a niche for this sort of thing for mobile delivery. But it's going to be short-lived. Once wireless broadband rolls out in larger metropolitan areas in a couple of years, you'll be able to listen to your favorite high-bandwidth streaming audio services--or even access the MP3 jukebox you have at home, for that matter. Of course, there will have to be a way to make up for the lost visual ad revenue for some streaming stations, but that will probably come from ISPs paying a blanket charge covering all of their subscribers, just as cable and satellite TV companies pay fees to providers of the "basic cable" channels.
Apologies. The Church and governments (often the same thing in those days) in Catholic countries put into effect a licensing system for all printing presses, and issued strict guidelines on what works could and couldn't be reprinted using the movable-type press. As a result, Protestant countries, particularly the liberal Netherlands, became the hubs for most publishing ventures early on, since they had the fewest restrictions and the lightest regulation.
The effect was similar. The most "valuable" documents to the pre-Gutenberg producers of books were precisely those books for which the Church had enjoyed a monopoly on the distribution.
Just as there were legal moves made to ban use of the movable-type printing press early on to protect the interests of scribes, here we're seeing the record industry try to use courts to artificially preserve a business and distribution model that flourished because the means of easy one-to-one distribution of music was inconvenient, expensive and difficult.
The broadband internet in general, and specialized peer-to-peer sharing tools like Napster are an incremental innovation, but a crucial one, just as the addition of movable type to the printing press was.
The notion of sales and replay royalties for recorded music is a relatively new one, dating back only to the piano roll and the gramophone in the second half of the 19th century. Before that, composers were paid lump sums for writing a new piece on commission, for performing, and in a few cases, for sales of sheet music. And this is what it will revert to: not a pay-per-download model, not a micropayment model, but no payment at all for recorded material. Not because people are selfish and have no respect for musicians and composers, but because the old means of enforecement--the inconvenience and high cost of copying--are vanishing.
The question is not whether the argument Boies is making here on behalf of Napster is going to win out, but when it will win out. The courts and the world's governments may try to preserve the status quo by law, but it's an unnatural state of affairs and will prove unenforceable within a matter of years.
Next to fall this way will be video as the price of bandwitdth and storage of peer-to-peer distributed video continues to drop. And finally, with the advent of personal binding technology and near-paper-quality flat screens and other display technology, we'll also see the end of enforceable copyright on print materials. The latter has had a respectable run of nearly 400 years, but this has only been because the quality of duplicates has been poor and the cost has been high. Once that's no longer the case--and it's just a few years away--so will go the last vestiges of the current print, sound and video revenue models.
The Gimp 1.0 series had an awful mess of a menu structure and insane dialog boxes. The current developer versions leading to 1.2, which are stable enough to use full time, have a vastly improved UI. Some folks who know UI design and commercial raster imaging packages have clearly been pitching in.
It's a shame GPL'ed mainstream-compliant color calibration probably can't be done without intellectual property lawsuits from Pantone. That's the big weakness I see remaining with GIMP.
Funny because the CIA and/or the New York Times thinks that drawing black boxes on top of the images in a PDF is "security" and sad because this does endanger the lives of the families of the people who collaborated with the CIA. When a few small business owners are killed or their houses are burned down in Los Angeles and Long Island in coming months, it will be because of this episode.
PDF is completely, utterly unprotectable. If you can read it, you can dissect it.
Have a "locked" PDF? Run it through Ghostscript, convert it back to Postscript, and do whatever you want with it.
Have a PDF encrypted for use with one of those "secure" book readers, as was done with the recent Stephen King novella? Then just run the reader alongside a debugger and intercept the Postscript parsing code to reconstruct the original file. Oh! But that's in violation of the licensiung terms of the "secure" reader!
The only way to make a PDF or Postscript file like this "safe" is a simple one. You need to replace the critical text or images with those black boxes, not simply cover it up using PDF or drawing tools. The original snippet of text needs to be utterly absent from the resulting file. Period. And then don't kid yourself if you think you can restrict or trace redistribution. Lock it all you want; the source Postscript can be recovered with a copy of Ghostscript and twenty minutes with the documentation, or failing that, by a C programmer armed with a debugger.
Last time I checked, Suns weren't overpriced. Their machines are very well engineered, with great I/O throughput, quality components, easy maintenance and upgrades, and responsive hardware support services. Their pricing isn't all that different from Compaq and IBM given the same quality hardware.
At the higher end--say, the 6000 series and up--you can hot swap and hot-plug CPUs. What Linux-friendly x86 vendors (or Linux distros, for that matter) support that?
Mom-and-pops and boutique vendors like VAResearch and Penguin Computing make cost-effective servers for the low to mid range.. with a constantly-shifting product line and component mix that drives engineers nuts. It's sometimes nice to be able to buy the same model configured the same way with the same components twice more than a year apart.
What are you comparing this pricing to? Dell's cheesy desktops in server cases?
Next question: if you're running a 100GB database for a $400 million company, would you put it on Linux, with a filesystem that will need 20 minutes to fsck in the event of an emergency reboot? Is "experimental" support for a shared fiber-channel disk array good enough to allow you to sleep soundly? Calling Sun or another "expensive" high-end vendor starts making more sense here.
Linux is forcing sun to beef up the services side of its business for revenue, as IBM has, and has hurt them--and everyone else--on the low end (1-4 CPU machines), for good reason. But where there's a big database or a heavy-lifting server application, nothing beats the so-called expensive stuff.
IBM's pretty much told everyone to clear off of OS/2 before the end of 2001. After that, even bugfixes will be something you'll need to contract IBM consultants for. It's highly unlikely you'll see any more major upgrades to Smartsuite or the Notes client for OS/2. Why should Sun do more than IBM?
At least you won't have to reformat the OS/2 machines to turn them into X terminals, right? Grab a big Linux or Solaris box and run StarOffice 5.2 remotely if you insist on being the last passenger on the ship.
OS/2 was nice. It was a better Windows than Windows for a while, and is still a better DOS than DOS--just ask your voice mail system vendor. But it's over. Time to pack up and move along. IBM is.
I'm more surprised IBM's pulled the plug on OfficeVision. Who will fill the demand for 5250 green-screen-terminal office suites? The humanity!
What's this nonsense about server farms and giant servers for 5,000 requests per second?
Your network traffic and the burden on the networking resources of your servers is the same regardless of whether you do push (ick) or pull. The key is in how hard you make the servers work to handle the requests.
What you want to do is cache the content and use dumb servers to push all the data out to clients. Something architected like the $100,000 Vignette Storyserver or the version of the free PHP FastTemplate package with caching added are the right idea, though you could do it with any reasonably lean server technology (in other words, CGI would be bad.
Presumably, you have a finite set of data snippets, many of which go to multiple users. So generate it once, store it on a filesystem, and leave it to one dumb, unthinking HTTP server and one or more proxy caches (like Squid or a commercial service like Akamai) in front of it to hand the static files out. Do your refresh, and so forth. If there's been no change in the data, don't update a given snippet file, and the servers' cache management and the web browser will only do a HEAD request and move on.
If you want to cut network traffic by adding bigger hardware, you could make your client request multiple snippets in a single request and use something as simple as server-side includes to concatenate the cached content snippets into a single returned file.
This sounds like a project that's generating revenue, either directly or indirectly. That should result in a budget for this.
My point wasn't that the Mac lacks a slick admin GUI, but that when you want to do serious administration of server elements running in the BSD layer of Mac OS X, it becomes (surprise!) Unix.
As an aside, Apple hardly has a monopoly on web-based server configuration. Have a gander at Liunuxconf, Solaris 8, any Cobalt product and so forth. And as far as Netinfo goes, multiplatform SNMP tools are quite a bit more widespread and have the added bonus of allowing you to manage a wide array of server applications on any number of operating systems, no longer just Unix.
Netinfo's interesting and nifty, but it's an eccentric cousin to LDAP and SNMP.
I never said Mac OS X wasn't pleasantly nifty. I do find it silly, however, to think it's going to storm the server world outside the same few hundred prepress shops running 100% MacOS networks.
Yeah, a Microsoft breakup will likely help Apple gain a bit more market share on the desktop and will make OS X for Intel a viable product, especially given its "fat binary" support for "hybrid" executable files.
But Mac OS X as a server OS has a key weakness: its nonstandard GUI. The easy, intuitive, and now stable GUI that Apple has put on top of OS X isn't built out of X Window.
It's a single-user GUI tied to local hardware despite the Unix running underneath. So at this point in time, remote administration of a Mac OS X machine needs to be done either with a destablizing, single-user remote control program like Timbuktu, or with the Unix command line. And not incidentally, OS X's BSD dialect is a pretty odd one, with a directory structure only an old NeXT-head from ten years back could love.
Furthermore, though Darwin, the non-graphical core of OS X, is open-source and free, OS X isn't. The most bug-prone, destablizing parts of OS X, which sysadmins raised on BSD and Linux would most want to be able to review and fix themselves, are closed and proprietary.
In addition, the Mac's well-deserved reputation for low-fuss plug-and-play hardware support comes largely from the Mac's closed, circumscribed world and its strictly limited selection of hardware. Putting Mac OS, whether the old one or OS X, on standard Intel hardware throws this out the window. Mac OS X will do no better at handling 700 disk controllers, 800 graphics chipsets, thousands of Ethernet cards and so forth than Windows and Linux do. And anyone who's spent much time with Macs lately knows that Apple's USB support is cranky and idiosyncratic to say the least, with vast numbers of devices that won't work off USB hubs or chained off the keyboard, even with external power.
About the only worthwhile insight in that silly little essay is that Mac OS X for Intel might be viable. Though unless Apple starts selling Intel hardware themselves, it's not likely to see the light of day, since Apple appears to be focused on making its money from hardware, not software: note the low price points for MacOS, AppleShare IP Server and now WebObjects. Netting $25 per copy for the sale of a boxed MacOS is a drop in the bucket once you factor in the cost of providing support.
This is insightful? KDE runs on all modern Unixes
on
David Faure Interview
·
· Score: 2
KDE isn't a "Linux" thing.. nor is GNOME for that matter. Precompiled, ready-to-run versions of KDE are available for several Unix variants including BSD, and it's long been available for Solaris as a one-step-install ".pkg" package.
As for running on NT.. NT isn't a Unix-like OS, so the entirety of KDE running on Win32 is fairly unlikely any time soon, even with the help of the porting and runtime libraries from companies like MKS. Windows isn't Unix, MacOS isn't Windows, and your refirgerator isn't a houseplant.
The Qt toolkit, on which KDE is built, however, is available for Windows--but not for free. Several of the apps in the KDE family are indeed available for Windows, and some others could easily be made available for it. But not so for apps that take advantage of the integration that the KDE layer provides. In other words, one might be able to port individual KOffice applications to Windows pretty easily, but they wouldn't integrate with each other, because the integration requires the KDE infrastructure underneath.
Similarly, the GTK+ toolkit, the basis of GNOME's UI, is also available under Windows. This has allowed GIMP to be ported to Windows; just don't expect it to communicate cleanly with other WIndows GTK+ apps like Mozilla without special work.
Seems to me, removing "non-free" is in keeping with Debian's mission. The practical reality is that it makes for a mediocre desktop environment without the non-free stuff, what with no RealPlayer and no release-quality full-featured browser.
I can see two good resolutions:
Leave it to the community to make derivatives of Debian, as Mandrake was (and still is, to some extent) to RedHat, which doesn't make much of a point, or...
Spin of "non-free" to outsiders, where it would remain available on outside FTP servers, and sell packaged versions that charge extra for "non-free", with all additional proceeds beyond production costs going to projects devoted to free replacements for key commercial packages
The real horror of recovery CDs
on
Copyrant
·
· Score: 2
Recovery CDs mean that a motherboard upgrade, or a change in key peripherals in a machine (say, new drive controllers or video cards introduced after the burn image's creation) will make a system recovery nigh impossible without buying entirely new licenses for the OS and for any apps bundled and tied into the recovery CD. Not so awful if you're using an "appliance"-style machine that has no swappable parts, like an iPaq. Very awful if you're using systems with standard ATX-style cases and motherboards.
It also means that a machine that's getting an OS version upgrade would need to be "recovered"--with the entire hard drive wiped--in order to end up with a reasonably clean, OS-rot-free install of the new OS version. As it is, it's already more cost effective in many cases to start fresh with a new full version of the latest OS when "recovering" a machine that's been upgraded from its original OS. That you cannot install an "upgrade" version from scratch by simply keying in the sequence of past OS license keys is part of the same greed. Microsoft clearly has long wanted it to be such a burden to "restore" a system that's been upgraded that customers would simply buy a new, full license to the latest OS rather than go through a multiple-step install.
Adobe, Macromedia and other companies with tedious, finicky license systems, are no doubt jealous. Their users would rebel if their installation schemes were quite this drastic. Ironically, their software is widely pirated by kids who can't afford it but want to learn it--and then go on to take jobs where they demand a copy. Has it ever occurred to them that the reason so many 22-year-old graphic designers are so passionate about--and competent in--Adobe's apps is that many have been using Adobe software since they were 14 years old?
For anything but a monopoly product like Windows or MS Office, "piracy" often benefits software vendors. This is why Oracle offers freely-downloadable, unrestricted copies of its core product line, as well as unrestricted CDs. Students, startups, developers and the curious can learn Oracle and the Oracle toolset, build a massive application, and test it on a pile of machines without paying a cent. But once they've gotten that far, at least in a country with enforcable copyright laws, they'll call in the Oracle reps and pay the $400,000 they owe before deploying.
No, this move to across-the-board recovery CDs only, which used to be the hallmark of low-end hardware vendors like Packard Bell, isn't about stopping piracy. It's about making system recovery and clean upgrades so difficult that more customers will opt to buy full versions of new OS releases for machines that already have a "license" because making use of that existing "copy" is too burdensome. Why sell upgrades to Win98, Win2K and Office for $90/$200/$240 when you can channel them into paying $200/$300/460 for convenience every time an upgrade is rolled out?
he only thing Proxy Server does that an ipchains-based setup won't is to forward HTTP packets based on the URL in them - i.e. you can have all requests for 'http://myorganisation.com/images/' sent to one server, and all requests from 'http://myorganisation.com/html/' sent to another server.
Squid will do this, and does it rather well if I recall. Should take about five minutes to configure the first time. Squid and ipchains server two very different purposes and can be used together without a hitch.
I use Bell Atlantic at home. I am at their mercy at work via third-party providers.
No matter who you go with in BA's service area, you're stuck with their incompetence when initially getting hooked up. There's not a thing you or anyone else can do about it. Nonetheless, you should by all means go with a different service provider. Getting connected is an absolute nightmare even if BA is your end-to-end provider because they're unserstaffed and undertrained at all ends of the business. BA's different groups don't coordinate with each other or even communicate regularly with each other. In practical terms, it's as though you're dealing with 3 different companies when getting connected. Don't make the mistake of thinking that going with BA simplifies things.
At least with a company like Flashcom, you have a shot at decent, responsive customer support after the initial nightmare is over. And you have a shot at getting something close to the advertised bandwidth, since you're not hanging entirely off of BA's overcrowded routers and hubs.
I'm sure you've seen this URL a few times already, but it can't be recommended enough. http://www.dslreports.com
Okay, I've been using Helix's tweaked GNOME 1.2 for a while now. The (non-mandatory) graphical installer was extremely impressive.
As for GNOME itself, Helix and their friends at Eazel and elsewhere have done a lot to make dialog boxes and menus consistent and rational, less of a homage to Emacs. Most apps still lack consistent keyboard accelerators across the UI, but they're getting there.
The control center, written largely by the RHAD folks, has also got a better, friendlier layout.. but its fundamental UI design borrowed from Linuxconf, of "swallowed" dialog boxes that you have to OK/Cancel out of before being allowed to switch cleanly to another panel, is still idiotic and has no precedent anyhere in the past 20 years of WIMP GUI design. Fix it. Now. Quite a few of the individual control center dialogs still can't hold a candle to MacOS, WIn98/Win2K or KDE in terms of clarity and ease of use.
The taskbar/panel is still fugly and still handles mouse events in nested menus in an uncivilized, amateurish manner. You need some subtle delays and stickiness, folks. By default.
Thank goodness for Sawmill/Sawfish. The window manager configuration is now swallowable by (and largely incorporated into) "GNOME" configuration. For non-geeks this is a Good Thing.
The Helix GNOME Updater is also slick and nice, very clear and friendly. But its UI is different from the Helix GNOME installer. As it should be. Except that Helix leaves you with no easy, friendly, non-geek way to install GNOME modules you passed on during the original install. For that, it's back to manual RPM or.deb installation or the powerful UI nightmare of GnoRPM or whatever your distro or OS makes you use. Boo, hiss.
Without Helix's enhancements and interface tweaks, GNOME 1.2 is a more stable, less bloated version of the same poorly-designed desktop environment we've had for a couple of years now. With Helix's tweaks, Miguel and crew have shown they understand that a well-thought-out GUI is something desirable. But there's still a lot of work left before GNOME will be fit for everyday use by non-techies. With 1.2, I can run GNOME now without too many problems, but I wouldn't wish it on friends or family. KDE, whatever else you might want to say about it, still has GNOME beat on usability hands-down.
IMAP's great. Server-side folders are key to any real email solution. Thing is, this makes an IMAP server (like a Notes server or an Exchamge server) more CPU- and disk- intensive. You can't support nearly as many people on the same hardware as you can with POP3.
Should you go with IMAP instead of a proprietary commercial maessaging system? Depends. If all you want is e-mail and adddressbooks, an IMAP server and an LDAP server will work just fine, and there are plenty of free web clients for remote access that you can chuck on top of any HTTP server.
But unless you're in an all *nix environment that can use iCal or the KDE or GNOME PIMs, if you plan to do group calendaring, scheduling, shared task lists and so forth, you're better off biting the bullet and going with something commercial, in which case I'd side with Domino/Notes mostly because they have a native Mac client for the non-email functionality, and a web inetrface that even extends to your custom apps, so you're not stuck with putting everyone who needs the scheduling but not on Win32 on feature-restricted webmail (contrast: Exchange). Also, you have a wide range of OSes you can run the server on (from Netware and WnNT to Linux, Solaris and OS/400). The Sun/Netscape/iPlanet suite is a political compromise, since the mail server is pure IMAP.. but as a groupware platform, it ain't no Notes.
Kick those IMAP zealots in the teeth unless they understand they're not going to be able to do any PIM-based, Palm-syncable group calendaring in a heterogeneous environment with Free Software. There are solutions, but no integrated or cross-platform ones. (Insert big asterisk for StarOffice, but that's another story).
That said, my experince has been that of the two free IMAP servers out there, Cyrus outperforms UW-IMAP. The latter puts each "mail folder" in a single mbox-style file, which is nice for POP-IMAP-Pine interoperability, but it also means crummy performance on large mailboxes, since an IMAP server has to "grep" for a full set of message headers every time a folder is accessed. Cyrus seems to do things more efficiently.
What OS do you want to run it on? Unless you're going to be supporting several thousand users or more, use anything you're comfortable with. Linux is plenty fine. BSD is fine. [Insert Un*x flavor here] is fine. If you've got a mainframe that can run it, that's fine too. NT would probably be okay, too, if you don't mind rebooting your mail server once a week or so to prevent memory leaks. If you're doing this for a large university or a large company with 10,000 or more users, the OS matters a bit more if you want to go with a few large servers rather than segmenting your mailboxes onto a bunch of smaller ones. Ah, politics.
Much more important than OS is the way you organize your disks and that you have enough RAM for the job. Assume 2-4MB of RAM will be in use for each user connected to the IMAP server, and since IMAP stays connected until told otherwise, that could mean quite a few people connected all day. As for disks, the faster the better, and if you have a separate drive (or striped drives) for the mail, all the better.
If you're in a mixed-OS environment, I'd go with a PAM-compliant Unix or with Linux. Both Cyrus and UW-IMAP can authenticate against pretty much anything, thanks to the wonder that is PAM. NIS? Fine. An NT domain via pam_smb? Fine. LDAP? Novell? Radius? Ditto.
-- it asked what country you were in. This -- was idiotic, given that the largest pool of -- visitors were in the US.
This isn't true; boo was highly advertised in Europe so had a far higher European visitor ratio than most.coms have.
Largest pool, I said. U.S. population: 300 million, with a higher level of internet penetration than any large European country. Surely Germany doesn't have as many internet users as the U.S. And Finland may be wired to the gills, but it's got fewer people than the Chicago metro area.
Show prices in Euros and dollars if you really must put Europatriotism over sales. Hell, once you're using so much Flash anyway, have a discreet map of the (NATO-vicinity) world in the corner of the page that zooms on rollover.. do something. Or do as others do. Have a discreet pointer for switching countries. Defaulting to the US may be offensive, but until there's a bigger single market for a B2C e-commerce site, it makes business sense. And I'm well aware that day is at hand, between the rollout of the Euro and Japan's mad rush to net adoption.
Anyway, Amazon doesn't seem to be having a problem with www.amazon.co.uk and www.amazon.de branding. Surely that's better than doing a lousy job of selling to everyone.
In my experience, documentation released under a restrictive NDA is distributed via physical media (CD, print, fax) after a signed NDA has been submitted. An anonymous clickwrap agreement, followed by a non-watermarked, unencrypted, unprotected PDF shows not even a cursory effort to protect the document from casual redistribution.
From a "real security" standpoint, there's not much of a practical difference between what they did and, say, distributing it on CD to a signer of an NDA. But this is almost as if they left the barn door open and then put a neon sign on the barn roof saying "OPEN BARN! TAKE OUR COWS!"
Slashdot Mirror
For an installation even one-tenth of this size, X terminals and big servers are the way to go. And no, the terminals don't need to be fast. We use P133s with 32MB just fine. Administering two, or even ten servers is a whole lot easier than administering 2500 independent workstations, even with the slickest use of NIS and NFS to pull off centralized configuration and management.
How much document sharing will there be with organizations that use other (read: Microsoft or Corel) office software? For interoperability and the shallowest learning curve for MS Office users, Corel WordPerfect Office and StarOffice are the best choices. Both will read existing MS Office files well enough, and will export new files in that format well enough, too. Neither is up to the task of heavy back-and-forth collaboration with MS Office users, but that's true of any office suite on any platform. StarOffice can't deal with most WordPerfect Office files. This would be a non-issue under 98% of circumstances, as so little of the world really uses WordPerfect anymore. However, as a state court system, you're part of the 2%: WordPerfect still has a strong presence at law offices, so interview a representative sample of users and managers to find out whether they currently do receive a notable number of WordPerfect files via e-mail or on disk. Frankly, they probably don't, instead getting them via fax or in hardcopy. So it may well not be an issue.
This answered and all other things being equal, I'd opt for StarOffice; Sun is in better financial shape these days than Corel, and bulky though StarOffice is, it's also natively written for Unix/Linux. It's also got more features that suit it for network and large-environment use. Both Corel and StarOffice are available in identical versions for Windows, so the laptop brigade can work seamlessly withe the terminal crowd.
For another thing, many job functions don't require an office suite at all. Don't assume everyone needs one, and don't just reflexively give one out, even if it's free of license fees like StarOffice. If someone just sends simple faxes and email, that's all they should be able to do. If someone simply accesses an AS/400 or mainframe and works with e-mail, they need access to nothing more than a web browser for e-mail (or perhaps Netscape Communicator with its IMAP mail support), and a tn5250 emulator. For sending faxes, use email-to-fax and fax-to-email gateways. Hylafax is your friend. And incidentally, StarOffice has nice hooks for printing and "emailing" through networked Hylafax servers. It can also sync with Palm gizmos.
For more complex environments, StarOffice has some further advantages. Enterprise-caliber support contracts are available, as are user and administration courses. Macros and scripts for it can be written in Javascript or in VBA. MS Office VBA scripts themselves won't work, but the skills some users and managers may have can be leveraged very easily. For another thing, it can be scripted with and interact with Java. So what? Ah, here's the nifty enterprise-caliber part: not only can StarOffice 5.2 access ODBC databases. It can also access any JDBC data source--which means pretty much any database on the planet, regardless of OS. In addition, you can take advantage of Java toolkits and SDKs for all sorts of things. For example, IBM has a toolkit for Java access for AS/400 client APIs. Want to add a menu item to StarOffice for retrieving data directly from a mainframe into a spreadsheet? Or linking calendar items (have I mentioned StarOffice's Outlook-like group calendaring?) dirtectly to AS/400 screens? You can. In ways that can be reused on other platforms and environments.
What Linux distribution you use is the least important piece of this. Choose something that offers easy creation of kickstart disks, to ease installation on new machines, and that offers good, reliable security upgrades. Me, I'd go with something that offers decent commercial support contracts. The terminals won't need any such nonsense, but it sure is nice to know you can get an engineer on the phone if a $50,000 server has a memory leak you can't squash. There are several ways to do this: you can go with a Linux vendor that offers contracts, like RedHat, or with a hardware vendor that sells Linux OS support on its hardware, like an IBM or VA Linux. For this reason, Mandrake comes out weaker than RedHat. It's not about the quality of the default installer or the number of "extras" on the CD. It's about maintainability going forward and the quality of support you can buy for those times when Usenet and online documentation cost too much downtime.
Depending on what you need in the way of file storage or backend applications, you may well want to look into a commercial Unix (say, Solaris, which runs StarOffice splendidly) on the backend. Linux is great, but if your needs really call for a SAN (and it doesn't sound like they do), or you want to go with one gigantic 24-CPU server instead of several 4-CPU ones, Linux may not cut it. Don't compromise your implementation for politics. Keep in mind that at this level, Linux is Unix is Unix, and that things like data and email and so forth can move fluidly between flavors without a moment's thought. Linux can certainly support 2500 users well; it can support tens of thousands of users well, as most large universities can tell you. There are also things it can do that something like Solaris can't, such as attach seamlessly to Windows file shares. But plan your applications and your network before you make the final OS decision, so the OS doesn't force compromises.
And another nice thing about these Linux X terminals is their flexibility. Just need green-screen VT102/3270/5250 access in the mailroom? Can do. Need to mix in some Windows-only applications after all? Set up a Metaframe server and give the X terminals access to an ICA client. Want users to be able to save to floppy, or attach a barcode reader? You can. With no changes on the terminals themselves.
Corel can't market their way out of a paper bag. They've got ugly ads, ugly packaging and lowball pricing that undermines confidence in the product.
Add to that a product strategy that seems to involve implementing features nobody ever asked for because they have orphaned technologies lying around, while leaving gnawing feature gaps intact year after year.
They're a slowly sinking ship. Their brand is synonymous with rescued orphanware, with shovelware, and with high-end software that only amateurs buy. It's one thing to be a niche vendor. It's quite another when the only niches you dominate are for word processing software for law offices, and graphic designers who have no money.
There's nothing special happening with Linux-related software reviews. Damn near all the reviews in the broad-audience print and web publications are superficial, usually on the positive side.
You need only read something by some of the exceptions like the New York Times's Peter Lewis or Byte's Jerry Pournelle to get a sense of proportion.
In the major computer media from IDG, ZD and CMP, reviews of office suites or $600 graphics tools seldom go beyond a checklist of features and a few comments on interface design. How often does it (StarOffice, MS Office) crash? How often does it (Adobe Illustrator 9) mangle files created with it to the point that they can't be opened?
Large-scale databases and high-end application servers sometimes get a proper review, with a realistic test environment, a significant amount of use, and real-world legacy data. But desktop operating systems and applications--even expensive ones, like design software and development tools--just get a review based on firing it up a couple of times and a look through the feature set.
Sure, that UML tool integrates with PVCS, but does it integrate well in an active team environmet? How's the interface? Does it freeze up when there is network latency? Yes, that desktop database can support tables with millikns of rows, but how did it perform on complex queries? And how did its "multiuser" features fare when you try that same query when three other users are making queries?
Linux-related reviews seem no better or worse. Caldera OpenDesktop installs easily? Great. How well preconfigured was Netscape? Did the PPP dialer setup utility require odd gymnastics like firing up a terminal window and running it as root from a command line? Did you have trouble accessing the update and patch download site? How responsive and knowledgable is the phone and email support staff you're paying for? Were tasks like printing a screenshot or installing new hardware easy or difficult?
If there is a general disconnect between the low polish level of many Linux desktop apps and the high praise Linux gets in the resulting review, that probably has something to do with the remarkable stability of Linux itself compared to the major consumer operating systems. No antialiased fonts? Awkward printer setup? But gee! The darn thing ran for four the weeks of the evaluation without a reboot.
It's tough to get a meaningful review of, say, Photoshop, from someone who is a product reviewer by trade, and not a full-time graphic artist. Even if the reviewer was an artist in a past job, s/he is seldom going to give the application the hard workout they would have in the course of real-world use. Instead, we get charts giving rendering times for specific effects, a database-style test suite applied to a creative tool, which should be more about flexibility, interface design and stability.
Enterprise applications are easier to review, since you can draw meaningful conclusions from a rigid, numerically quantifiable test suite. But even here, reviewers grab onto the superficial to make major pronouncements. So-and-so's web-based server management console is lacking? But what if most users of the application use a command line to manage the product anyway, as with many databases and web and application servers?
Pirhana is part of the standard RH 6.2 distribution. What you're getting here is a year of support; but more key to this, you're also getting consulting services: For your $2000, they will provide, via phone and email, the exact configuration information necessary to set up a node's worth of clustering on your network.
Comfortable with documentation and understand clustering already? Then get the $150 SSL-enhanced version of 6.2 and set it up yourself. Everything you need is there.
Don't need an RSA-licensed Apache-SSL in the node? Then grab Redhat 6.2 for free and configure Pirhana (which is included) at no cost.
Consulting costs money. If you don't need consulting, good for you. High-availability clustering is free.
So there's another company trying to compete with DMX, MusicChoice, Muzak and SkyRadio. Their one hook is that they're trying to be the first to get portable and mobile receivers made, which will go thr way of the dodo, quadraphonics, 8-track and AM stereo if the receivers are designed only to hook to their system.
Satellite and cable narrowcast "radio" stations have been around for some years now. And yes, while DMX and Sky are available to digital cable and home sattelite customers at home, their core revenue source is commercial subscribers: shops, restaurants, offices, buildings and so forth, which pay a higher fee.
There's certainly a niche for this sort of thing for mobile delivery. But it's going to be short-lived. Once wireless broadband rolls out in larger metropolitan areas in a couple of years, you'll be able to listen to your favorite high-bandwidth streaming audio services--or even access the MP3 jukebox you have at home, for that matter. Of course, there will have to be a way to make up for the lost visual ad revenue for some streaming stations, but that will probably come from ISPs paying a blanket charge covering all of their subscribers, just as cable and satellite TV companies pay fees to providers of the "basic cable" channels.
Apologies. The Church and governments (often the same thing in those days) in Catholic countries put into effect a licensing system for all printing presses, and issued strict guidelines on what works could and couldn't be reprinted using the movable-type press. As a result, Protestant countries, particularly the liberal Netherlands, became the hubs for most publishing ventures early on, since they had the fewest restrictions and the lightest regulation.
The effect was similar. The most "valuable" documents to the pre-Gutenberg producers of books were precisely those books for which the Church had enjoyed a monopoly on the distribution.
Just as there were legal moves made to ban use of the movable-type printing press early on to protect the interests of scribes, here we're seeing the record industry try to use courts to artificially preserve a business and distribution model that flourished because the means of easy one-to-one distribution of music was inconvenient, expensive and difficult.
The broadband internet in general, and specialized peer-to-peer sharing tools like Napster are an incremental innovation, but a crucial one, just as the addition of movable type to the printing press was.
The notion of sales and replay royalties for recorded music is a relatively new one, dating back only to the piano roll and the gramophone in the second half of the 19th century. Before that, composers were paid lump sums for writing a new piece on commission, for performing, and in a few cases, for sales of sheet music. And this is what it will revert to: not a pay-per-download model, not a micropayment model, but no payment at all for recorded material. Not because people are selfish and have no respect for musicians and composers, but because the old means of enforecement--the inconvenience and high cost of copying--are vanishing.
The question is not whether the argument Boies is making here on behalf of Napster is going to win out, but when it will win out. The courts and the world's governments may try to preserve the status quo by law, but it's an unnatural state of affairs and will prove unenforceable within a matter of years.
Next to fall this way will be video as the price of bandwitdth and storage of peer-to-peer distributed video continues to drop. And finally, with the advent of personal binding technology and near-paper-quality flat screens and other display technology, we'll also see the end of enforceable copyright on print materials. The latter has had a respectable run of nearly 400 years, but this has only been because the quality of duplicates has been poor and the cost has been high. Once that's no longer the case--and it's just a few years away--so will go the last vestiges of the current print, sound and video revenue models.
The Gimp 1.0 series had an awful mess of a menu structure and insane dialog boxes. The current developer versions leading to 1.2, which are stable enough to use full time, have a vastly improved UI. Some folks who know UI design and commercial raster imaging packages have clearly been pitching in.
It's a shame GPL'ed mainstream-compliant color calibration probably can't be done without intellectual property lawsuits from Pantone. That's the big weakness I see remaining with GIMP.
Funny because the CIA and/or the New York Times thinks that drawing black boxes on top of the images in a PDF is "security" and sad because this does endanger the lives of the families of the people who collaborated with the CIA. When a few small business owners are killed or their houses are burned down in Los Angeles and Long Island in coming months, it will be because of this episode.
PDF is completely, utterly unprotectable. If you can read it, you can dissect it.
Have a "locked" PDF? Run it through Ghostscript, convert it back to Postscript, and do whatever you want with it.
Have a PDF encrypted for use with one of those "secure" book readers, as was done with the recent Stephen King novella? Then just run the reader alongside a debugger and intercept the Postscript parsing code to reconstruct the original file. Oh! But that's in violation of the licensiung terms of the "secure" reader!
The only way to make a PDF or Postscript file like this "safe" is a simple one. You need to replace the critical text or images with those black boxes, not simply cover it up using PDF or drawing tools. The original snippet of text needs to be utterly absent from the resulting file. Period. And then don't kid yourself if you think you can restrict or trace redistribution. Lock it all you want; the source Postscript can be recovered with a copy of Ghostscript and twenty minutes with the documentation, or failing that, by a C programmer armed with a debugger.
Last time I checked, Suns weren't overpriced. Their machines are very well engineered, with great I/O throughput, quality components, easy maintenance and upgrades, and responsive hardware support services. Their pricing isn't all that different from Compaq and IBM given the same quality hardware.
At the higher end--say, the 6000 series and up--you can hot swap and hot-plug CPUs. What Linux-friendly x86 vendors (or Linux distros, for that matter) support that?
Mom-and-pops and boutique vendors like VAResearch and Penguin Computing make cost-effective servers for the low to mid range.. with a constantly-shifting product line and component mix that drives engineers nuts. It's sometimes nice to be able to buy the same model configured the same way with the same components twice more than a year apart.
What are you comparing this pricing to? Dell's cheesy desktops in server cases?
Next question: if you're running a 100GB database for a $400 million company, would you put it on Linux, with a filesystem that will need 20 minutes to fsck in the event of an emergency reboot? Is "experimental" support for a shared fiber-channel disk array good enough to allow you to sleep soundly? Calling Sun or another "expensive" high-end vendor starts making more sense here.
Linux is forcing sun to beef up the services side of its business for revenue, as IBM has, and has hurt them--and everyone else--on the low end (1-4 CPU machines), for good reason. But where there's a big database or a heavy-lifting server application, nothing beats the so-called expensive stuff.
IBM's pretty much told everyone to clear off of OS/2 before the end of 2001. After that, even bugfixes will be something you'll need to contract IBM consultants for. It's highly unlikely you'll see any more major upgrades to Smartsuite or the Notes client for OS/2. Why should Sun do more than IBM?
At least you won't have to reformat the OS/2 machines to turn them into X terminals, right? Grab a big Linux or Solaris box and run StarOffice 5.2 remotely if you insist on being the last passenger on the ship.
OS/2 was nice. It was a better Windows than Windows for a while, and is still a better DOS than DOS--just ask your voice mail system vendor. But it's over. Time to pack up and move along. IBM is.
I'm more surprised IBM's pulled the plug on OfficeVision. Who will fill the demand for 5250 green-screen-terminal office suites? The humanity!
Unix and Linux StarOffice 5.x installs fine as a multiuser application. You just need to read the instructions.
/net argument.
In order to install it multiuser, you have to be logged in as root and start the installer from the command line with the
What's this nonsense about server farms and giant servers for 5,000 requests per second?
Your network traffic and the burden on the networking resources of your servers is the same regardless of whether you do push (ick) or pull. The key is in how hard you make the servers work to handle the requests.
What you want to do is cache the content and use dumb servers to push all the data out to clients. Something architected like the $100,000 Vignette Storyserver or the version of the free PHP FastTemplate package with caching added are the right idea, though you could do it with any reasonably lean server technology (in other words, CGI would be bad.
Presumably, you have a finite set of data snippets, many of which go to multiple users. So generate it once, store it on a filesystem, and leave it to one dumb, unthinking HTTP server and one or more proxy caches (like Squid or a commercial service like Akamai) in front of it to hand the static files out. Do your refresh, and so forth. If there's been no change in the data, don't update a given snippet file, and the servers' cache management and the web browser will only do a HEAD request and move on.
If you want to cut network traffic by adding bigger hardware, you could make your client request multiple snippets in a single request and use something as simple as server-side includes to concatenate the cached content snippets into a single returned file.
This sounds like a project that's generating revenue, either directly or indirectly. That should result in a budget for this.
The bulk packs include black, white, green, gray and red.
No yellow, no blue. How are you supposed to build a full-scale Lego replica of an IKEA store without yellow and medium blue?
My point wasn't that the Mac lacks a slick admin GUI, but that when you want to do serious administration of server elements running in the BSD layer of Mac OS X, it becomes (surprise!) Unix.
As an aside, Apple hardly has a monopoly on web-based server configuration. Have a gander at Liunuxconf, Solaris 8, any Cobalt product and so forth. And as far as Netinfo goes, multiplatform SNMP tools are quite a bit more widespread and have the added bonus of allowing you to manage a wide array of server applications on any number of operating systems, no longer just Unix.
Netinfo's interesting and nifty, but it's an eccentric cousin to LDAP and SNMP.
I never said Mac OS X wasn't pleasantly nifty. I do find it silly, however, to think it's going to storm the server world outside the same few hundred prepress shops running 100% MacOS networks.
Yeah, a Microsoft breakup will likely help Apple gain a bit more market share on the desktop and will make OS X for Intel a viable product, especially given its "fat binary" support for "hybrid" executable files.
But Mac OS X as a server OS has a key weakness: its nonstandard GUI. The easy, intuitive, and now stable GUI that Apple has put on top of OS X isn't built out of X Window.
It's a single-user GUI tied to local hardware despite the Unix running underneath. So at this point in time, remote administration of a Mac OS X machine needs to be done either with a destablizing, single-user remote control program like Timbuktu, or with the Unix command line. And not incidentally, OS X's BSD dialect is a pretty odd one, with a directory structure only an old NeXT-head from ten years back could love.
Furthermore, though Darwin, the non-graphical core of OS X, is open-source and free, OS X isn't. The most bug-prone, destablizing parts of OS X, which sysadmins raised on BSD and Linux would most want to be able to review and fix themselves, are closed and proprietary.
In addition, the Mac's well-deserved reputation for low-fuss plug-and-play hardware support comes largely from the Mac's closed, circumscribed world and its strictly limited selection of hardware. Putting Mac OS, whether the old one or OS X, on standard Intel hardware throws this out the window. Mac OS X will do no better at handling 700 disk controllers, 800 graphics chipsets, thousands of Ethernet cards and so forth than Windows and Linux do. And anyone who's spent much time with Macs lately knows that Apple's USB support is cranky and idiosyncratic to say the least, with vast numbers of devices that won't work off USB hubs or chained off the keyboard, even with external power.
About the only worthwhile insight in that silly little essay is that Mac OS X for Intel might be viable. Though unless Apple starts selling Intel hardware themselves, it's not likely to see the light of day, since Apple appears to be focused on making its money from hardware, not software: note the low price points for MacOS, AppleShare IP Server and now WebObjects. Netting $25 per copy for the sale of a boxed MacOS is a drop in the bucket once you factor in the cost of providing support.
KDE isn't a "Linux" thing.. nor is GNOME for that matter. Precompiled, ready-to-run versions of KDE are available for several Unix variants including BSD, and it's long been available for Solaris as a one-step-install ".pkg" package.
As for running on NT.. NT isn't a Unix-like OS, so the entirety of KDE running on Win32 is fairly unlikely any time soon, even with the help of the porting and runtime libraries from companies like MKS. Windows isn't Unix, MacOS isn't Windows, and your refirgerator isn't a houseplant.
The Qt toolkit, on which KDE is built, however, is available for Windows--but not for free. Several of the apps in the KDE family are indeed available for Windows, and some others could easily be made available for it. But not so for apps that take advantage of the integration that the KDE layer provides. In other words, one might be able to port individual KOffice applications to Windows pretty easily, but they wouldn't integrate with each other, because the integration requires the KDE infrastructure underneath.
Similarly, the GTK+ toolkit, the basis of GNOME's UI, is also available under Windows. This has allowed GIMP to be ported to Windows; just don't expect it to communicate cleanly with other WIndows GTK+ apps like Mozilla without special work.
I can see two good resolutions:
Recovery CDs mean that a motherboard upgrade, or a change in key peripherals in a machine (say, new drive controllers or video cards introduced after the burn image's creation) will make a system recovery nigh impossible without buying entirely new licenses for the OS and for any apps bundled and tied into the recovery CD. Not so awful if you're using an "appliance"-style machine that has no swappable parts, like an iPaq. Very awful if you're using systems with standard ATX-style cases and motherboards.
It also means that a machine that's getting an OS version upgrade would need to be "recovered"--with the entire hard drive wiped--in order to end up with a reasonably clean, OS-rot-free install of the new OS version. As it is, it's already more cost effective in many cases to start fresh with a new full version of the latest OS when "recovering" a machine that's been upgraded from its original OS. That you cannot install an "upgrade" version from scratch by simply keying in the sequence of past OS license keys is part of the same greed. Microsoft clearly has long wanted it to be such a burden to "restore" a system that's been upgraded that customers would simply buy a new, full license to the latest OS rather than go through a multiple-step install.
Adobe, Macromedia and other companies with tedious, finicky license systems, are no doubt jealous. Their users would rebel if their installation schemes were quite this drastic. Ironically, their software is widely pirated by kids who can't afford it but want to learn it--and then go on to take jobs where they demand a copy. Has it ever occurred to them that the reason so many 22-year-old graphic designers are so passionate about--and competent in--Adobe's apps is that many have been using Adobe software since they were 14 years old?
For anything but a monopoly product like Windows or MS Office, "piracy" often benefits software vendors. This is why Oracle offers freely-downloadable, unrestricted copies of its core product line, as well as unrestricted CDs. Students, startups, developers and the curious can learn Oracle and the Oracle toolset, build a massive application, and test it on a pile of machines without paying a cent. But once they've gotten that far, at least in a country with enforcable copyright laws, they'll call in the Oracle reps and pay the $400,000 they owe before deploying.
No, this move to across-the-board recovery CDs only, which used to be the hallmark of low-end hardware vendors like Packard Bell, isn't about stopping piracy. It's about making system recovery and clean upgrades so difficult that more customers will opt to buy full versions of new OS releases for machines that already have a "license" because making use of that existing "copy" is too burdensome. Why sell upgrades to Win98, Win2K and Office for $90/$200/$240 when you can channel them into paying $200/$300/460 for convenience every time an upgrade is rolled out?
he only thing Proxy Server does that an ipchains-based setup won't is to forward HTTP packets based on the URL in them - i.e. you can have all requests for 'http://myorganisation.com/images/' sent to one server, and all requests from 'http://myorganisation.com/html/' sent to another server.
Squid will do this, and does it rather well if I recall. Should take about five minutes to configure the first time. Squid and ipchains server two very different purposes and can be used together without a hitch.
I use Bell Atlantic at home. I am at their mercy at work via third-party providers.
No matter who you go with in BA's service area, you're stuck with their incompetence when initially getting hooked up. There's not a thing you or anyone else can do about it. Nonetheless, you should by all means go with a different service provider. Getting connected is an absolute nightmare even if BA is your end-to-end provider because they're unserstaffed and undertrained at all ends of the business. BA's different groups don't coordinate with each other or even communicate regularly with each other. In practical terms, it's as though you're dealing with 3 different companies when getting connected. Don't make the mistake of thinking that going with BA simplifies things.
At least with a company like Flashcom, you have a shot at decent, responsive customer support after the initial nightmare is over. And you have a shot at getting something close to the advertised bandwidth, since you're not hanging entirely off of BA's overcrowded routers and hubs.
I'm sure you've seen this URL a few times already, but it can't be recommended enough. http://www.dslreports.com
Okay, I've been using Helix's tweaked GNOME 1.2 for a while now. The (non-mandatory) graphical installer was extremely impressive.
.deb installation or the powerful UI nightmare of GnoRPM or whatever your distro or OS makes you use. Boo, hiss.
As for GNOME itself, Helix and their friends at Eazel and elsewhere have done a lot to make dialog boxes and menus consistent and rational, less of a homage to Emacs. Most apps still lack consistent keyboard accelerators across the UI, but they're getting there.
The control center, written largely by the RHAD folks, has also got a better, friendlier layout.. but its fundamental UI design borrowed from Linuxconf, of "swallowed" dialog boxes that you have to OK/Cancel out of before being allowed to switch cleanly to another panel, is still idiotic and has no precedent anyhere in the past 20 years of WIMP GUI design. Fix it. Now. Quite a few of the individual control center dialogs still can't hold a candle to MacOS, WIn98/Win2K or KDE in terms of clarity and ease of use.
The taskbar/panel is still fugly and still handles mouse events in nested menus in an uncivilized, amateurish manner. You need some subtle delays and stickiness, folks. By default.
Thank goodness for Sawmill/Sawfish. The window manager configuration is now swallowable by (and largely incorporated into) "GNOME" configuration. For non-geeks this is a Good Thing.
The Helix GNOME Updater is also slick and nice, very clear and friendly. But its UI is different from the Helix GNOME installer. As it should be. Except that Helix leaves you with no easy, friendly, non-geek way to install GNOME modules you passed on during the original install. For that, it's back to manual RPM or
Without Helix's enhancements and interface tweaks, GNOME 1.2 is a more stable, less bloated version of the same poorly-designed desktop environment we've had for a couple of years now. With Helix's tweaks, Miguel and crew have shown they understand that a well-thought-out GUI is something desirable. But there's still a lot of work left before GNOME will be fit for everyday use by non-techies. With 1.2, I can run GNOME now without too many problems, but I wouldn't wish it on friends or family. KDE, whatever else you might want to say about it, still has GNOME beat on usability hands-down.
IMAP's great. Server-side folders are key to any real email solution. Thing is, this makes an IMAP server (like a Notes server or an Exchamge server) more CPU- and disk- intensive. You can't support nearly as many people on the same hardware as you can with POP3.
Should you go with IMAP instead of a proprietary commercial maessaging system? Depends. If all you want is e-mail and adddressbooks, an IMAP server and an LDAP server will work just fine, and there are plenty of free web clients for remote access that you can chuck on top of any HTTP server.
But unless you're in an all *nix environment that can use iCal or the KDE or GNOME PIMs, if you plan to do group calendaring, scheduling, shared task lists and so forth, you're better off biting the bullet and going with something commercial, in which case I'd side with Domino/Notes mostly because they have a native Mac client for the non-email functionality, and a web inetrface that even extends to your custom apps, so you're not stuck with putting everyone who needs the scheduling but not on Win32 on feature-restricted webmail (contrast: Exchange). Also, you have a wide range of OSes you can run the server on (from Netware and WnNT to Linux, Solaris and OS/400). The Sun/Netscape/iPlanet suite is a political compromise, since the mail server is pure IMAP.. but as a groupware platform, it ain't no Notes.
Kick those IMAP zealots in the teeth unless they understand they're not going to be able to do any PIM-based, Palm-syncable group calendaring in a heterogeneous environment with Free Software. There are solutions, but no integrated or cross-platform ones. (Insert big asterisk for StarOffice, but that's another story).
That said, my experince has been that of the two free IMAP servers out there, Cyrus outperforms UW-IMAP. The latter puts each "mail folder" in a single mbox-style file, which is nice for POP-IMAP-Pine interoperability, but it also means crummy performance on large mailboxes, since an IMAP server has to "grep" for a full set of message headers every time a folder is accessed. Cyrus seems to do things more efficiently.
What OS do you want to run it on? Unless you're going to be supporting several thousand users or more, use anything you're comfortable with. Linux is plenty fine. BSD is fine. [Insert Un*x flavor here] is fine. If you've got a mainframe that can run it, that's fine too. NT would probably be okay, too, if you don't mind rebooting your mail server once a week or so to prevent memory leaks. If you're doing this for a large university or a large company with 10,000 or more users, the OS matters a bit more if you want to go with a few large servers rather than segmenting your mailboxes onto a bunch of smaller ones. Ah, politics.
Much more important than OS is the way you organize your disks and that you have enough RAM for the job. Assume 2-4MB of RAM will be in use for each user connected to the IMAP server, and since IMAP stays connected until told otherwise, that could mean quite a few people connected all day. As for disks, the faster the better, and if you have a separate drive (or striped drives) for the mail, all the better.
If you're in a mixed-OS environment, I'd go with a PAM-compliant Unix or with Linux. Both Cyrus and UW-IMAP can authenticate against pretty much anything, thanks to the wonder that is PAM. NIS? Fine. An NT domain via pam_smb? Fine. LDAP? Novell? Radius? Ditto.
-- it asked what country you were in. This
.coms have.
-- was idiotic, given that the largest pool of
-- visitors were in the US.
This isn't true; boo was highly advertised in Europe so had a far higher European visitor ratio than most
Largest pool, I said. U.S. population: 300 million, with a higher level of internet penetration than any large European country. Surely Germany doesn't have as many internet users as the U.S. And Finland may be wired to the gills, but it's got fewer people than the Chicago metro area.
Show prices in Euros and dollars if you really must put Europatriotism over sales. Hell, once you're using so much Flash anyway, have a discreet map of the (NATO-vicinity) world in the corner of the page that zooms on rollover.. do something. Or do as others do. Have a discreet pointer for switching countries. Defaulting to the US may be offensive, but until there's a bigger single market for a B2C e-commerce site, it makes business sense. And I'm well aware that day is at hand, between the rollout of the Euro and Japan's mad rush to net adoption.
Anyway, Amazon doesn't seem to be having a problem with www.amazon.co.uk and www.amazon.de branding. Surely that's better than doing a lousy job of selling to everyone.
In my experience, documentation released under a restrictive NDA is distributed via physical media (CD, print, fax) after a signed NDA has been submitted. An anonymous clickwrap agreement, followed by a non-watermarked, unencrypted, unprotected PDF shows not even a cursory effort to protect the document from casual redistribution.
From a "real security" standpoint, there's not much of a practical difference between what they did and, say, distributing it on CD to a signer of an NDA. But this is almost as if they left the barn door open and then put a neon sign on the barn roof saying "OPEN BARN! TAKE OUR COWS!"