Linked by Thom Holwerda on Wed 9th Jan 2013 10:33 UTC, submitted by Straylight
BeOS & Derivatives Ars 'reviews' Haiku, and concludes that "at the end of the day, Haiku may not be much more than an interesting diversion, something to play with on a spare bit of hardware on a rainy afternoon just for a bit of fun. But even if it amounts to no more than that, Haiku is still worth checking out." The article is a bit scant on content, but it does give me the opportunity to link to my review of Haiku alpha 1 from 3 years ago. I try Haiku every now and then to see if that review needs an update, but it always amounts to 'it got a bit more stable' - which is fantastic, but not a reason to redo it.
Thread beginning with comment 547961
To read all comments associated with this story, please click here.
Member since:

The BeBox was released in October 1995.

No, it was "announced" in October 1995, the first 100 developer boxes didn't ship till 1996.

It was a curious beast, sporting dual 66MHz PowerPC 603e processors,

Or 133MHz... the 133's are the ones still worth having. The 66's are horribly underpowered.

...and concentrated on selling the operating system by itself, initially for PowerPC Macintoshes.

This implies to me it was just a magic drop in OS for the Mac, when clearly it wasn't. It was a hard slog. But, the Mac port was well established by the time hardware production was dropped. Mac's ran a version of DR8, after all. I had a copy of PR1 that would boot on Mac or BeBox.

While negotiating a sale price, Gassée, exuding his typical bombastic confidence, told a reporter that “we’ve got Apple by the balls and we’re going to squeeze until it hurts.” Someone at Apple got wind of this and phoned a guy, who phoned another guy...

That's very apologetic with regards to BeOS... truth was, at the time the OS didn't meet the specs and Openstep was far closer. BeOS was also hampered by some of the internal code - given it was either licensed from a third party or shady GPL usage.

You can also never underestimate Steve Job's power of sales, once he got wind of an opportunity.

BeOS reached its pinnacle of success in 2000 when the R5 version was released as a free download.

Yes - personal version. With giant restrictions. Like not supposedly being able to be installed to a real partition (as that was actually a selling point of Pro.) The fact that droves of users used the partition install loophole was only a minor part of the reasons Be Inc failed. Because...

... and a last-ditch attempt to save the company by bundling BeOS with the Sony eVilla Internet Appliance

NO! Total fantasy. Thom and I are now screaming, "FOCUS SHIFT!!"

What really happened:

Web Appliances became a buzzword, everyone circa 1999 - 2001 was trying to launch some kind of net enabled appliance - be it a semi computer or tablet or toaster. Be saw this, decided to change focus and create BeIA, BeOS for Internet Appliances. Except, that failed dismally because no average consumers actually wanted the overpriced underpowered boxes and Sony bailed on BeIA almost before the eVilla was released. Compaq did the same thing with the Cliper. And the WebPads all later shipped with WinCE.

Secondly, BeIA is NOT BeOS. BeOS is NOT BeIA. BeIA shares a lot of the core OS with BeOS, but it is as like BeOS in reality as iOS is to MacOS X. There's a compatible API in the earlier builds (I've personally seen a lot of pre BeIA 1.0 builds running, and played with the dev kit.) BeIA did a lot different. It had a compressed file system (CFS), it had a special kernel that allowed ELF executables to be compressed (CEL format), which then used a master symbol table to uncompress the exe to run it. It was a bunch of extra drivers, such as touch screen and wireless. I personally owned a DT300 webpad and it ran BeIA from a 16MB CF card (yes MB, not GB.) The browser was the main app - Wagner (an Opera 4 based monstrosity) and there were COM like abilities and various cool comms features in the underlying OS (i.e. Binder.) Yes, one could trich the OS in to booting to a severely limited and very crippled Tracker, but it didn't really work very well and wasn't all that useful.

I'm not going to go on, but just that alone made me sceptical of the rest of the article.

Reply Score: 6

Thom_Holwerda Member since:

That's very apologetic with regards to BeOS... truth was, at the time the OS didn't meet the specs and Openstep was far closer. BeOS was also hampered by some of the internal code - given it was either licensed from a third party or shady GPL usage.

Wait, what?

Reply Parent Score: 1

henderson101 Member since:

;-) <-- this is all you are getting on this forum.

Reply Parent Score: 3

MYOB Member since:

A number of network drivers, PCMCIA support (likely not around at the time of the Apple possible purchase though) and similar were either lifted wholesale from Linux with some vague code releases that didn't always compile and rarely gave compatible binaries; or licenced from their original copyright holders for the Linux version.

There was also Intel licenced code in the USB stack (again, later than the Apple era), Metrowerks licenced IDE, licenced MIDI stack, the 3D rendered, some other drivers were written and bought in from external sources.

What Be could have sold was about 80% of an OS with a lot of licencing issues to sort out.

Reply Parent Score: 3

MOS6510 Member since:

The BeOS Personal Version was part of my multi-boot system featuring 6 operating systems. I seem to recall that to boot BeOS I needed a boot floppy. The rest could be booted from a boot menu.

It seemed like a cool idea, until you realize you don't use most operating systems and they are just a waste of space.

I think it was BeOS, MS-DOS, Windows 98, Windows 2000, Linux and ehm... another one, not *BSD. Ah, I think I counted Windows 3.1 as the 6th, it started from the MS-DOS install.

Reply Parent Score: 2