Slashdot Mirror

That was his point, don't you think?
Wasting 30 seconds searching would have given you http://simplex.giss.nasa.gov/s..., or http://www.cesm.ucar.edu/model... or http://www.mi.uni-hamburg.de/S... ... and many, many more.

Funny thing, the code, the data, the explanations, everything has been avalable for years, and yet so many of the public believe they're not. I wonder why that is?

It's like there was this massive political campaign against science. Of which you just became part of. Congratulations!

"They make a point now of not sharing the details of the models with people. That would concern you if you had any intellectual curiosity."

Well, that is bullshit of incredible intensity. Just for curiosity, who exactly told you that? You presumably won't have any injection to sharing that data with us.
In return, I will share with you this thing called Google, with which I was able quite rapidly to find:
http://www.gfdl.noaa.gov/fms
http://mitgcm.org/public/sourc...
http://www.giss.nasa.gov/tools...
http://www.nemo-ocean.eu/About...
http://www.metoffice.gov.uk/re...
http://forge.ipsl.jussieu.fr/i...
http://www.giss.nasa.gov/tools...
http://www.cesm.ucar.edu/model...
http://www.cesm.ucar.edu/model...
http://www.cesm.ucar.edu/model...
http://edgcm.columbia.edu/
http://www.mi.uni-hamburg.de/f...
http://www.mi.uni-hamburg.de/H...
There's more but I'm tired of cut and pasting. You would be able to find these also if you have any intellectual curiosity, but them you might have to doubt the sources of your info on how bad the climatology people are, and how they're hiding the code to conceal that out doesn't work, and that maybe the models do run and give outputs, ave before you know it the foundations of you whole world view are shaken.

"They make a point now of not sharing the details of the models with people. That would concern you if you had any intellectual curiosity."

Well, that is bullshit of incredible intensity. Just for curiosity, who exactly told you that? You presumably won't have any injection to sharing that data with us.
In return, I will share with you this thing called Google, with which I was able quite rapidly to find:
http://www.gfdl.noaa.gov/fms
http://mitgcm.org/public/sourc...
http://www.giss.nasa.gov/tools...
http://www.nemo-ocean.eu/About...
http://www.metoffice.gov.uk/re...
http://forge.ipsl.jussieu.fr/i...
http://www.giss.nasa.gov/tools...
http://www.cesm.ucar.edu/model...
http://www.cesm.ucar.edu/model...
http://www.cesm.ucar.edu/model...
http://edgcm.columbia.edu/
http://www.mi.uni-hamburg.de/f...
http://www.mi.uni-hamburg.de/H...
There's more but I'm tired of cut and pasting. You would be able to find these also if you have any intellectual curiosity, but them you might have to doubt the sources of your info on how bad the climatology people are, and how they're hiding the code to conceal that out doesn't work, and that maybe the models do run and give outputs, ave before you know it the foundations of you whole world view are shaken.

My links didn't seem to format properly, so here they are in un-edited format : http://cat.inist.fr/?aModele=afficheN&cpsidt=13671279 http://onlinelibrary.wiley.com/doi/10.1111/j.1559-1816.2003.tb01876.x/abstract

These both clearly show social facilitation for cognitive tasks, with effects that increase as the task is complexified.

The article you put forth, by the way, doesn't support your initial postulate that the Ringelmann effect "means" the group cannot perform better than the sum of its parts. They did not disprove in any way the Köhler effect. They merely were not able to say that the variation was statistically significant. This is very interesting (particularly given that the second study, concerning word associations, indicates a decrease in performance in almost all group situations), but I can't find a follow-up work. Usually, this is the sign that there was a parameter that was unaccounted for and after correcting the structure of the experiment the results were "as expected", or merely that upon repeating the experiment they observed the expected outcome. Furthermore, I can't seem to find the paper you are putting forth in its full form, nor in a journal. The work in question seems to have been presented at two working groups (the EAESP General meeting 2005 and the Jena Social Psychology Task Group 2005), and then quite suddely vanished. The University hasn't kept the presentations (the wayback machine has archived them, but the links on the university's own page are dead : http://epb.uni-hamburg.de/de/node/3703 ), Dr Nina Plum now works as a consultant, and I can't find Gabriele Engelhardt on either the university's website nor an open search. Dr. Witte has retired, but only in 2011, and after 2005 there does not seem to be any collaboration between any of the three authors to deepen the elements presented in this research.

This isn't an attempt at character assassination, but I feel you're engaging in what would pass for irrational behaviour in conspiracy circles. Evidence that goes your way is necessarily true, despite being incomplete, not peer reviewed, nor having been followed-up, whereas evidence that suggests the opposite is "wrong" despite meeting all these criteria.

viper - http://www.informatik.uni-hamburg.de/RZ/software/emacs/viper/viper_3.html

A small piece of me weeps when I see this.

VHDL Cookbook is a good, though dated, intro.

Use ghdl to learn vhdl, without the need to have hardware, as it compiles VHDL to an executable. Icarus is similar, but for Verilog. gEDA has good tools, including the gtkwave waveform viewer. Combined, ghdl, Icarus and gtkwave are a pretty useful simulation suite. You can go a long way with simulation, since the normal design flow is to get the system 100% using simulation, then as a last step program the FPGA with maximal probability of it just working. As Bruce said, the actual partition, place and route tools are proprietary and specific to each FPGA vendor, and a google search will come up with a number of cheap FPGA boards.

Keep an eye on left field though. There is a convergence in progress between desktop CPU's, GPU's, parallel systems and FPGAs (which can be seen as an array of massively parallel simple processors). One day all I wrote may be obsolete and you will be able to program your FPGA in CUDA, or whatever results when mainstream programming figures out how to handle parallel systems properly.

VHDL Cookbook is a good, though dated, intro.

Use ghdl to learn vhdl, without the need to have hardware, as it compiles VHDL to an executable. Icarus is similar, but for Verilog. gEDA has good tools, including the gtkwave waveform viewer. Combined, ghdl, Icarus and gtkwave are a pretty useful simulation suite. You can go a long way with simulation, since the normal design flow is to get the system 100% using simulation, then as a last step program the FPGA with maximal probability of it just working. As Bruce said, the actual partition, place and route tools are proprietary and specific to each FPGA vendor, and a google search will come up with a number of cheap FPGA boards.

Keep an eye on left field though. There is a convergence in progress between desktop CPU's, GPU's, parallel systems and FPGAs (which can be seen as an array of massively parallel simple processors). One day all I wrote may be obsolete and you will be able to program your FPGA in CUDA, or whatever results when mainstream programming figures out how to handle parallel systems properly.

There's lots of climate-model source-code available on the web. Much of it has been publicly available for years.
Examples:

http://www.giss.nasa.gov/tools/modelE/
http://www.ccsm.ucar.edu/
http://www.mi.uni-hamburg.de/Projekte.209.0.html?&L=3
http://www.sciencemag.org/cgi/content/full/317/5846/1866d/DC1
http://geoflop.uchicago.edu/forecast/docs/Projects/modtran.html

Now for all the skeptics out there -- those of you who have downloaded and tested any climate code, submitted patches, constructive suggestions, etc. to the code developers, please stand up and give us a shout-out!

Don't be shy or modest -- even if you've done nothing more than submit a one-line change to a makefile, let's hear about it!

The 1980's called, they want their software security model back.

This was already implemented with the BSG 9 virus on the Amiga.
http://agn-www.informatik.uni-hamburg.de/catalog/amiga/html/bgs9terr.htm

Xix.

I looked up kernel package and here is the screenshot! http://www.biologie.uni-hamburg.de/b-online/library/maize/www.ag.iastate.edu/departments/agronomy/images/corn/fig49.gif

This does not affect the users directly, but it is a major pain for integrators/porters. OO.o has a terrible habit of bundling all of the 3rd-party software packages, that it uses, into its own source tree. I'm talking about (probably missed some):

• agg
• bash
• bitstream-vera
• bsh
• bison
• boost
• curl
• db42
• dmake
• expat2
• freetype
• icu
• jpeg
• firefox (or some other Mozilla-based browser)
• libmspack
• libsndfile
• libtextcat
• libwpd
• libxslt
• neon
• nss
• nspr
• python
• sane-backends
• STLport
• unixODBC
• unzip
• vigra
• xmlsec1
• xt
• zip
• zlib

If they could, I'm certain, they would've bundled Java too, but — fortunately — Sun's license prohibits that... Now I realize, that this is done to offer "a single package" to those, who build it on their own, but nobody does. Everybody gets these from their OS' integrators. And the pain for us is enormous, because to force OO.o build to stop its silly ways is a serious undertaking. For some of the above packages there is --with-system-foo configure-flag, but not for all, and the default is to always use the bundled one, so support for the external ones bitrots quickly...

Most of the local builds don't bother and so end up wasting disk space and CPU-time rebuilding packages, which are external to OO.o. The end results are also bloated, duplicating stuff, that's already installed on the users' systems and without bug-fixes, which have already gone into each of the respective package since its most recent "bundling" into OO.o tarballs.

Download a source tarball and see for yourself... Something like: tar tjf OOo_OOG680_m9_source.tar.bz2 | grep 'z$'. No other software project does this on this scale and for good reasons — it is Just Wrong[TM]. OO.o better clean up their act in this respect...

Go and play a little with this tool. Choose a fixed amount of defects and play with chip dimensions.

Plants do not require a full spectrum.
As can be seen in this picture, plants absorb at specific frequencies (which you might have guessed by the fact that they are green)

So you could use fluorescent lights focused around the useful wavelengths. Cutting out the infra-red and ultraviolet that isn't useful. You also don't need the same intensity as sunlight because you have light all the time. No pesky clouds or nights.
NASA uses LEDs to grow plants for experimentation in space.

Such fluorescent lights aren't really produced right now. Mainly because the only people who would want to use them to grow pot.

emacs has viper-mode, if that's what you're talking about.

Everything DNA related, shares a common encoding scheme. The following web page explains the basic of amino acid encoding. DNA consists of four nucleotides A, T, C and G. For proteins, triplets of these are used to specify any one of 24 amino acids used. But this could simply be the DNA equivalent of Logo or VHDL (a programming language used to specify silicon chip circuits).

For some organs of the body to grow into complex shapes, some cells have to be pre-programmed to die at the correct time in order for folding to take place. So this has to be encoded in the DNA.

DNA could also be used as the tape of a Turing Machine storing permanent state information (such as which genes are active in a honey bee).

then use viper

Adam McNaughton wrote a quick and hilarious synopsis folk song about Shakespeare's Hamlet... verses here:

http://www.informatik.uni-hamburg.de/~zierke/marti n.carthy/songs/oorhamlet.html

You call Monkey.B "the good old days"? I remember the Lamer Exterminator!

I don't agree. Data (aka assertion) that is not explicit fact, is an interpretation of explicitly stated facts. NLP tries to label each part of a sentence into nouns, verbs, adjective, pronouns, adverbs and so on. It's goal is to form a graph that describes what the sentence means. The NLP term for the relationship is valence. Here is a random paper from google on the topic http://nats-www.informatik.uni-hamburg.de/~ingo/pa pers/tal2000.pdf.gz

RDF Graph (not RDF schema) attempts to describe data in a way that computers don't have to "figure out" (ie parse) the subject predicate object. This is achieved through writing rules about a given topic. Dependency grammar "can be considered" techniques and frameworks for representing and generating knowledge. NLP also uses rules to process/parse sentences. In the case of NLP, it's grammatical and contextual rules.

The two topics are much closer and have much more common than superficial similarities. One difference is that RDF does not address the issue of building knowledge through automated process. Humans have to do it. People are lazy. Ignoring the problem doesn't make the problem go away.

I agree there's a lot more is needed to achieve the goal of semantic web, but the question is, will W3C listen to others and change? So far it hasn't.

Hey! The University of Hamburg is quite a prestigious university, you insensitive clod!

An interesting (and excellent) link on VHDL coding standards in a working environment is also available off that page: the European Space Agency's VHDL coding standard (available in PostScript format here http://tech-www.informatik.uni-hamburg.de/vhdl/doc /style_guide/ModelGuide.ps.gz).

The Xilinx WebPack is a great place to start - you get everything you need to take you from text-edited files to a binary image on one CD (or download from here: http://support.xilinx.com/support/download.htm). It even comes with a (very) cut-down version of Mentor Graphics' ModelSim and Xilinx's own synthesis tool, XST.

If you use Altera chips, they have a similar offering, called Quartus II Web Edition (http://www.altera.com/products/software/products/ quartus2web/sof-quarwebmain.html)

Speaking of text editors, (X)Emacs has a great VHDL mode that can beautify your code, create makefiles and manage your projects, available here:http://opensource.ethz.ch/emacs/vhdl-mode.htm l.

An interesting (and excellent) link on VHDL coding standards in a working environment is also available off that page: the European Space Agency's VHDL coding standard (available in PostScript format here http://tech-www.informatik.uni-hamburg.de/vhdl/doc /style_guide/ModelGuide.ps.gz).

The Xilinx WebPack is a great place to start - you get everything you need to take you from text-edited files to a binary image on one CD (or download from here: http://support.xilinx.com/support/download.htm). It even comes with a (very) cut-down version of Mentor Graphics' ModelSim and Xilinx's own synthesis tool, XST.

If you use Altera chips, they have a similar offering, called Quartus II Web Edition (http://www.altera.com/products/software/products/ quartus2web/sof-quarwebmain.html)

Speaking of text editors, (X)Emacs has a great VHDL mode that can beautify your code, create makefiles and manage your projects, available here:http://opensource.ethz.ch/emacs/vhdl-mode.htm l.

If you're learing about VHDL, consider writing some code for OpenCores as part of your education. I gather they have server problems at the moment, so the link may not work.

This debate is old. The vitalists were a group who believed that life contained some special essence. Most of what I have read about this historical controversy comes from Ernst Mayr. All attempts to identify processes that rely on a vital force failed. All identified processes follow the known rules of biochemistry and thermodynamics. There does not seem to be any "spark of life," and this debate died out in mainstream biology in the face of the repeated successes of physical approaches.

here

A much larger version of the SIGGRAPH `94 image "Photon Soup", clocking in at 840x560, can be found HERE.

That is, until the PC gets hit by a ton of viruses, spyware, adware, trojans... which slow it down to half the speed of the Mac.

Had a PC since 1990 - never got a virus. A little prevention goes a long way.

You know OSX people are not invulnerable to viruses or trojans. I remember the Morris Worm infected mostly BSD. The current relatively safe OSX situation occurs because with with 2% market share it's not worth the ego boost for most virus kiddies to target the platform. Basically, when it comes to Macs no kiddies really care... or cared.

However, I recall in the late 1980s when Macs had quite a large market share relative to today that there were several Mac-specific viruses. But personally I'd say it's just a matter of time before OSX gets its own virus. I'd actually see the release of an OSX-specific virus or trojan as evidence of increasing market share!

Do a test with ATI Radeon 9200 and XFree 4.4 (You can use a farily new distro with back patches for XFree 4.3). The XFree 4.4 drivers for ATI Radeon 9200 is not made by ATI. They are native OpenSource XFree drivers made with documentation provided by ATI. Most if not all Radeons up to 9200 has good native support with full 3d accelration. The tv-out is not as good as it should, but it's been worked on.

"Gatos - Enhanced XFree86 drivers for ATI videocards, with emphasis on TV-input and Output"

Radeon XFree4.3 manual page

atitvout 0.4 - This program is mostly a student hack, but it shows how the Radeon techicaly works regarding tv-out.

In the Operating Systems Comparison section, ESR fails to take note of the Lisp Machine. It's instructive to note that Lisp Machine hackers were the main contributors to the Unix-Haters Handbook and the Unix-Haters mailing list.

> The litigious nature of this society is drawing it into a very frightening pattern of litigating for profit.

Only the details are new... they used to hide in a bush until you came walking down the road, at which point they jumped out waving their sword and crying "Half your pack belongs to me!", and you could either draw your own sword or else hand over the goods without a fight.

Now we use lawyers instead of swords, but the basic concept hasn't changed. And there's a potential for lots of profit, if you're good enough with your sword and the shire reeve doesn't string you up.

In case anyone is interested...

If you inserted a disk where the "disk blocks in use" bitmap was corrupt (it had a checksum), Workbench would usefully repair it by scanning all existing files on disk and writing a new block usage map, before allowing you to write files to the disk. To use a UNIX analogy, the Amiga would fsck floppys before mounting them.

However, the 256Kb Kickstart ROM did not have room for this disk repair code. So, the disk repair code was loaded from disk. Unfortunately, the first place Workbench tried was on the corrupt disk itself, the executable code ":L/Disk-Validator".

The SADDAM virus (it was released during Gulf War 1) used this as an initial infection vector. It was initially spread on a damaged disk. Upon inserting the disk, Workbench 1.3 would automatically load and run :L/Disk-Validator, which contained the virus code. Other than this exploit, there is no way a disk can force the Amiga to load and run something just through disk insertion.

The SADDAM infection code would create a :L/Disk-Validator file containing the virus on any write-enabled floppy disk that was inserted, and would deliberately invalidate the bitmap again after every disk write operation finished. There were numerous clones of this virus.

It became completely obsolete in Workbench 2.0 (released in 1990, available as standard in low-budget Amigas in 1991/1992), as the 512Kb Kickstart ROMs had plenty of space for the Disk Validator code, so it was no longer loaded from disk.

* Wil Wright invents the "software toy" or "sandbox" type of game with SimCity


The invention of the "software toy" is *very* important - but there were earlier examples:

1982 Rock'y Boots

1960's Logo

1960's Eliza and Parry

I'm sure there are earlie examples of interactive, exploratory, non-games.

In fact the gimp pretty much needs a pretty major rearchitecture to get it where it needs to be. This is being done in GEGL, the core rewrite for Gimp 2.

The main feature is that of arbitrary color spaces (12/16 bit components, floating point components, CYMK etc ) . From what I can tell, they've implemented thier own template-alike system for acheiving the generic programming techniques this requires. Very efficient polymorphic behaviour almost demands compile time stuff - its just not practical to go through literally millions of function calls when processing a large image. AFIACS, gegl uses a preprocessor to achieve this. The work doesn't appear to be going incredibly fast.

It would have been more interesting to see this written in a non app specific language that supports generic programming like C++ - have a look at Vigra - no i in there!
It might be even more interesting, if perhaps restricting to the developer pool, to implement the image processing in an efficiently compilable functional language like Haskell or ML.

I agree with the comments about the gimp interface. I was shocked and saddened when sodipodi adopted the abomination as well... poor gnome.

For the original radeon and close derivatives (radeon mobility M6, M7, mobility 7500, and IGP 320M/340M but not regular radeon 7500 or anything above), the way to get TV output is to boot the computer up with the TV plugged in, and run atitvout. This program clones the normal display onto the TV. All acceleration features that are normally supported on the display are also supported on the TV.

For radeon 7500 and above, a different technique is required: there you simply boot up the computer with the TV attached, and TV output is automatic. But, if you want to run X on the TV output you must use the VESA driver and not the radeon driver. Consequently the card's acceleration features are not available in X on the TV output. However, as a special case, video playback overlay acceleration is available on the TV out using the xvidix driver in mplayer.

Basically, the situation is not ideal, but it is a long way from being hopeless.

Huh? Why would you want to do that - Symbolics Lisp machines already had an excellent editing substrate called EINE (EINE Is Not Emacs) and later ZWEI (Zwei Was EINE Initially). Zmacs, the LispM editor, was using those and is possibly still more advanced than the Emacs of today.

You can take a look at The Symbolics LispM Museum for more information and pictures.

A mirror of the document is here.

And here is the master thesis of Tom Knight, describing the architecture of the Lisp Machine. If you want to see one in action, visit me on the Chaos Communication Camp.

One online Symbolics Lisp Machine museum is here.

And yes, UNIX royally sucks. It plays in the same suckage leage as Windows, of course, but still it sucks. It's a clone of technologies of the early 70ies, and a bad one.

You should check out ifplugd. It's a daemon that automaticially configures your network device when a cable is plugged into it, and unconfigures it when the cable is unplugged.

I don't believe it currently works with all network cards, but it does work on many of them (read, works fine in my laptop)

Even early versions of LISP were their own OS

See here for extensive information about dedicated Lisp machines.

I guess Mr. Flat (what a name!) hasn't ever heard of the LISP Machine..

I am a Java programmer myself (laugh it up), but Swing just plain annoys me.

I like the Swing APIs when viewed from the application developer side, but the underlying implementation is hopeless baroquen. Swing was the first GUI API to get my attention since Genera on Symbolics LISPms.

I am a Java programmer myself (laugh it up), but Swing just plain annoys me.

I like the Swing APIs when viewed from the application developer side, but the underlying implementation is hopeless baroquen. Swing was the first GUI API to get my attention since Genera on Symbolics LISPms.

While I agree that MacOS X has lots of the features, it doesn't have 'em all.

It's removable media handling seems a bit clunky compared to the amiga. While the GUI presents quite an Amiga-like view, the underlying OS, exposed through the command line, tells a somewhat different story.

Also, in my tests anyway , most operating systems get a bit confused when I "encourage" a CD out of one drive, while it has files open on it, and insert it into an externally connected USB CDROM drive. Amigas handle basically equivalent operations relatively well, usually just leaving the process in a wait state on the open file (though some disk write operations can still get a bit upset), and telling you to put the CDROM back in a drive (any CDROM drive, not necessarily the one you took it out of). This is one of those things where people from other platforms just say "don't do that" or, even worse "most people only have one CDROM drive" - they've got so used to the arbitrary restriction of not removing in-use media that they can't get a handle on a system where that restriction does not exist. Imagine watching a video file from CD, and popping the disk out. The video just pauses. You stick another CD in, and copy something from it to your harddrive. You stick the video CD back in. The video resumes. Or maybe you realise that you started watching the video from your cd-writer, and now you want to burn a CD, so you take the video cd out of your cd-writer, and put it in your other cdrom drive. Now substitute "near-arbitrary file operations" for "watching a video"

Also MacOS X's CLI is nothing like CLIM. It's basically just like having an XTerm open on an X desktop. Or an AmigaShell on Workbench. Or even a CMD.EXE window open on WinNT. CLIM is like the AutoCAD UI, generalised.

Have a read through this to get a better idea about what I mean about a CLIM-like GUI/CLI combo. It's also similar to AutoCAD.

Additionally, you noone should hope for lotus-effect-coated cars too soon. The automobile industry has been searching for "anything-repelling" coatings for ages now and one of the counterparts (apart from car washers losing their jobs) is that you simply cannot coat repelling coatings, so if you have to repair the coating after an accident, this simply won't work (or will be very expensive). There will certainly be a use for this technique, but it'll probably not be as obvious as most people think.

Be sure to study the Lisp de-facto standard UI system, CLIM 2.0, closely. It's great for complicated GUI systems for OO Lisp projects.

Alsom, be aware that the best GUIs tend to be the ones that let the user do most anything your underlying program code can, and combine things in unexpected ways. Make sure to read up on Naked Objects

So how much better (apart from being sterilized) is this compared to the stuff you can buy in the supermarket (whose particles are spherical to some degree)?

I do not believe you are correct. In a modern processor, almost NO power is consumed when computations are not happening. CMOS logic is set up so that as long as its state does not change, there is no current flowing (well, transistors "leak", but that is relatively little current). However, when the clock ticks, all the gates start changing states. During the state change, power is consumed. This page has some excellent applets which show how and when the current flows.

During computation is the only time power is being used. Because of how CMOS works, power usage grows approximatekly linearly with clock speed - which makes sense - if power is consumed when stuff changes state, doubling the clock doubles the number of state changes per time, increasing power consumption.

Common Lisp has a lot of benefits for this type of work. Since it is completely dynamic (ie - everything runs in an image with which you can interact, add code/compile and debug, all at run-time), the plug-in/scripting is taken care of from the start, and can have the full syntax of CL and access to any of the main program's features you choose to give it. CL will probably give you the most results per line/minute of code because of this dynamism.

Most CL implementations have pretty good foreign function interfaces for C and C++ libraries (Franz's Allegro CL even provides support for run-time Java objects.)

CL's performance is on par with C++ in general, and lags only in one major area - FP operations require "boxing" overhead when the symbol pointing to the numbers is dynamically typed (most compilers optimize statically typed declarations pretty well - which makes most of the overhead go away.)

Of course, before you go off on your great quest, you should probably read what some of the other posters have suggested. Writing graphics software like the type you describe is an incredible amount of work (I gave up my uber-Scheme system after 100 lines and settled for writing smaller utilities and plugins), and many have tried and miserably failed before.

• "The Symbolics-Lisp system software constitutes a large-scale programming environment, with over a half-million lines of system code accessible to the user. Object-oriented programming techniques are used throughout the Symbolics-Lisp system to provide a reliable and extensible integrated environment without the usual division between an operating system and programming languages. All of the system software is written in Symbolics-Lisp."

There you see the basic concepts of Brix and Crush. Symbolics had that in 1984. One of the Symbolics people wrote a post-mortem,"The Lisp Machine: Noble Experiment or Fabulous Failure?", which explains what's wrong with this concept better than I could.

• "The Symbolics-Lisp system software constitutes a large-scale programming environment, with over a half-million lines of system code accessible to the user. Object-oriented programming techniques are used throughout the Symbolics-Lisp system to provide a reliable and extensible integrated environment without the usual division between an operating system and programming languages. All of the system software is written in Symbolics-Lisp."

There you see the basic concepts of Brix and Crush. Symbolics had that in 1984. One of the Symbolics people wrote a post-mortem,"The Lisp Machine: Noble Experiment or Fabulous Failure?", which explains what's wrong with this concept better than I could.