Ramblings about the Smalltalk Language and environments that I've had experiences with. Warning: may include small parts of other musings.
19 November 2019
Pharo Takeover
Well, a small group of far-left authoritarians have taken over the Pharo Project. They've single-handedly instituted a "Code of Conduct" promulgated by an individual that hates the heterosexual world, trying to force other people to accept and condone the the bizarre trans world. There a small faction that ascribes to the far-left world view, and they're trying to force all the others to accept their rule. I've cancelled my contributions to the Pharo Association and to some projects I was supporting. Good luck to them, Pharo is persona non-grata as far as I'm concerned. I abandoned Squeak Smalltalk because they didn't accept contributions except people in the clique. It looks like Pharo has decided to go down the same path. Long live normal Smalltalk, not controlled by a radical minority.
01 September 2019
Something different
Just though I'd plug two other blogs I write, Object Networks Blog(the company blog) and Objectnets Open Source. I write about almost anything programming related in theses blogs, I don't just stick to Smalltalk topics. I hope you enjoy them, and maybe learn something new.
22 April 2019
Hurray!!
Hey I finally got some feedback from my postings!
Mariano Martinez Peck said:
Hi John, I was just checking your sites and blog and it looks pretty cool and what you have been doing with single boards. I will stay tunned! I am personally interested in SoB , IoT, etc and here are Instantiations we also want to continue targeting that more and more. You can check yourself about that in mine, Seth Berman, and Instantiations official twitter and linkedin everything we have been doing.
So far we worked with Raspberry in particular. We added GPIO accessing and MQTT binding, and we already had since several years, minimal images (server running in 1.5MB), remote debugging, stack dumping (similar to Fuel stack dump) and many other features that are quite nice for IoT. (Some stuff redacted here…) We have a long list of devices to try and probe the portability of our VM.
Anyway, if you are interested on learning more, lets us know.
Best,
And I replied:
I have to update my website (https://objectnets.net and https://objectnets.org) because I also experiment with Arduino stuff (Uno, Due, and Mega 2560). In fact, I just spent 4 1/2 hours last night playing with the Uno. I must have several hundred buried in IoT gear, not even considering the ancillary gear I been collecting over the decades. I’d love to do more work with Smalltalk on these devices, if I only had more time…
14 August 2015
Back again
There’s an interesting web site called SMALLTALK ZEN about Web Development with Pharo Smalltalk and Seaside. Pharo Smalltalk contains the reference implementation of Seaside, a web framework that makes it easier to develop web applications. There’s an article on Wikipedia about Seaside and then there’s the Seaside web site.
30 December 2006
Snow
I thought that the 8 inches of snow we got in Albuquerque last week was something, but now we have 16 inches of snow that came over the past two days. The previous snowfall had mostly melted; the temperatures have been warm enough to keep the snow from piling up.
At least we can use the precipitation. Hope it melts soon, because we may be getting more on Tuesday.
At least we can use the precipitation. Hope it melts soon, because we may be getting more on Tuesday.
29 December 2006
GNU Smalltalk
I've seen several misconceptions about the way GNU Smalltalk works. This post includes a few of them. Yeah, I know the post is from a while ago.
Gnu Smalltalk by default is not graphic based.
The weird file syntax is one from the fileIn format. The bang (exclamation point) is used to "separate logical items". This format is also used by Squeak. Just fire up the "file list" browser in Squeak (available from the Open... menu item) to access the same type of files.
Adding the Browser package includes the gui most people are used to in Smalltalk. Adding the browser to GST is accomplished by starting GST and executing the following:
PackageLoader fileInPackage: 'Browser' !This loads the package that you're interested in and takes care of the dependencies, if any. You can then save the image if you want so that you don't have to reload the pacakges with
ObjectMemeory snapshot: '/some/file/name' !
27 December 2006
Gap
There's been a pretty big gap in my postings. I've been busy with other things, including a move.
05 March 2006
Verizon Redux
Category: Mobile Phones
I just noticed this disclaimer on Verizon Wireless' site:
Unlimited NationalAccess/BroadbandAccess services cannot be used (1) for uploading, downloading or streaming of movies, music or games, (2) with server devices or with host computer applications, including, but not limited to, Web camera posts or broadcasts, automatic data feeds, Voice over IP (VoIP), automated machine-to-machine connections, or peer-to-peer (P2P) file sharing, or (3) as a substitute or backup for private lines or dedicated data connections.
Isn't that service useful. I wonder if they'll provide a useful consumer product. Probably not because of the RIAA and the DMCA.
Blog phones
Category: Mobile Phones
Here's a pointer to an interesting development, Sony Ericsson teams with Google. Pictures of the phones here. Blogging from a mobile phone, pictures included. That could be a useful product, considering that I use a blog to note things interesting to me, if nobody else.To be released 2nd quarter 2006. I wonder who's going to offer them in the US. One phone operates on
EDGE networks I've heard of, but not in my area. EDGE technology is based on GSM. that's what the G is in EDGE. Cingular Wireless and T-Mobile have EDGE networks in the United States.
I looked up UMTS and it seemes to be EDGE for CDMA networks. It's also known as WCDMA and is based on CDMA technology and was envisioned for the next generation of GSM. It's a European standard designed to support data transmission rates of 144kbps for use in vehicles, 384kbps for pedestrian use and up to 2mbps for use indoors. I found this info here. I haven't found a U.S. carrier that supports it. Back in 2001, Verizon Wireless seemed to be moving in that direction, from what I've been researching.
I found this info on high speed data networks (mobile) on the Verizon site. They seem to have two internet services, Wireless Internet NationalAccess and BroadbandAccess. NationalAccess operates on a laptop at typical speeds of 60-80 kbps and bursts up to 144 kbps while Broadband Access operates at average speeds of 400 – 700 kbps. Other than that, no other info is available (that I can find).
26 February 2006
More Silt 4
Well, still more problems. James Robertson created a new Silt pre-built with the latest changes. I downloaded it and installed it on the Mac. When I went to http://localhost:9000/blog/BlogView I got this:

Grrrrrrr. I'd really like to use Silt for a multi-user multi-blog set for a non-profit that I belong to but I don't have the time to debug the problems I'm encountering with Silt. Oh well.
Yet another foolish quiz
| You Are an Espresso |
At your best, you are: straight shooting, ambitious, and energetic At your worst, you are: anxious and high strung You drink coffee when: anytime you're not sleeping Your caffeine addiction level: high |
What Kind of Coffee Are You?
You drink coffee when: anytime you're not sleeping
Your caffeine addiction level: high
Well, that's definitely true
Your caffeine addiction level: high
Well, that's definitely true
22 February 2006
More Silt 3
Trying this over again. When I tried to reuse the imageSiltFromRepository everything appeared to be broken. The new dirctory is imageSiltFR2. This image is built with SiltSSPFiles 1.26, Silt 2.25 and Client-Core 6.1. I need to check to see if these are the correct version numbers.
After setting up blog1 and blog2, I tried to go to and I got a bizarre error relating to '/usr/local/vw7.3.1nc/imageSiltFR2/blog2/View.ssp/css/impact-blog.css'. There's no such file, of course. There is a file '/usr/local/vw7.3.1nc/imageSiltFR2/source/css/impact-blog.css'. Hmmm.
Oh no, not me!
Well, there was a posting on the FRIAM mailing list that pointed to some nerd/geek tests. The posting welcomed participants to take the test, so I did. These are the results.
I don't think that the tests go into enough detail to ascertain the true nerdiness/geekiness of the test taker; but maybe it's just me :^)
Oh, by the way, the best periodic table I found was at www.dayah.com/periodic/ (hey, it didn't say that you couldn't use aids, did it?).
I don't think that the tests go into enough detail to ascertain the true nerdiness/geekiness of the test taker; but maybe it's just me :^)
Oh, by the way, the best periodic table I found was at www.dayah.com/periodic/ (hey, it didn't say that you couldn't use aids, did it?).
17 February 2006
More Silt 2
I worked on the Silt from the Repository again. This time I'm tried to set up more blogs to see what happens to the blogs when I set up the values differently in the Blog Manager. So far I've set up the blogs jp2, blog3 and blog4. The blog3 and blog4 work OK. The blog jp2 is probably screwed up because the DES encryption module doesn't work and I haven't learned to just terminate it.
10 February 2006
More Silt
Given the problems I found previously with the Silt blog server pre-built module, I decided to move the deployment over to a local box. It's going to be easier to work out the bugs here rather than a remote box.
After having the same problems locally as I had remotely, I've also decided to try the download from Repository option.
The repository download seems to work better than the pre-built Silt. It comes up, I can post and I can set up multiple blogs.
The setup for mutiple blogs has some problems though. There's no feedback from the blog setup mechanism, so you can't tell if it worked. Worse, it seems to keep adding layers of directories, so that I've gotten a url like this http://localhost/blogname/
09 February 2006
Siltier, or the fallout from installing Silt
Well, I tried to install the blog server called Silt, written by James Robertson of Cincom Smalltalk fame. I first tried the pre-built image and configuration that is supposed to be self-contained except for your VW 7.3.1 VM.
I found that the thing didn't work unless you had all 3 of the XML parcels from the goodies/other directory and the WebToolkit from the web directory. I did this because I wanted to install Silt on a webserver and I didn't like carrying the whole VW 7.3.1 distribution (~500MB) onto the webserver. As it is the Silt distro is 31 MB with the total space used at 91 MB. I have other uses for the extra 400MB.
Well, I got all the extra modules set up on the server, did the
export VISUALWORKS=visualworks/home/directory
started the server, go to http://www.objectnets.com:9000/blog/blogView and I get this message:
Application Error
A fatal error has occurred in this application. Please contact this site's administrator.
No such file or directory ("/home/blogs/public_html/silt/image/blog/blogView")
My first thought is that the "No such file or directory..." should be in the log for the web server, not displayed on the error page.
Continued...
06 January 2006
Not all Smalltalks are the same
This post should have a post date of December 12, 2005, but I haven't had a chance to update my blog until now.
In this blog(?) posting The Power of Smalltalk IDE's Talios seems to mistake Dolphin IDE for all Smalltalk environments. I don't know exactly what he seems to be commenting on, because I've not seen the latest Dolphin Smalltalk (version 6). One of his problems is the autocomplete feature of Dolphin which seems to be referred to as "intellisense completion". Having run into this feature in other IDEs, I can appreciate his frustrations. I don't exactly type too fast (I'm being gentle with myself, my typing speed and accuracy is abysmal) and that feature is more of a bane than a blessing, and if not implemented well it's just a horror show.
06 December 2005
He hasn't a queue
James Robertson has an interesting and telling post in his blog,
We have a failure of Imagination, which has a reply to Elliotte Rusty Harold's Today's News, although my problems with the post are different than James'.
Java's
List class does not lack any of the functionality in Ruby's. Java just factors it out into a few more classes, especially the Collections class, and skips a couple of rarely used "convenience" methods. The result is a simpler, easier-to-understand, easier-to-use, more humane API.I'd hardly call the feeble classes in Java "humane", maybe "less intimidating" for the type of programmers that use Java. Programmers that have had some time using Java end up with large libraries of "methods" that work, and that fill in the holes in the Java libraries.
A bigger problem that I have with Elliotte Rusty Harold's post is that he makes some comments that I find astonishing.
Another example: Fowler likes the
first and last methods in Ruby, but list.first() is not significantly simpler than list.get(0). list.last() is perhaps a little simpler than list.get(list.size() - 1) but only because Java stupidly indexes everything from 0 rather than 1.Well, other than
list.get(0).list.last() is considerably uglier and less obvious than list.first(), but then again, you can always document what you were trying to do so that some other programmer can understand the code.But this is the comment that really takes the cake.
And how often do you actually need to get the first item in the list? Needing the last item in a list is even less common.
Is he kidding? Has he taken a Data Structures and Algorithms course? I'd recommend that he get the Data Structures book by Aho, et al. It's right on Amazon. The second chapter, "Basic Data Types", has sections on "The Data Type, List", "Implementation of Lists", "Stacks", "Queues", "Mappings", and "Stacks and Recursive Procedures". Since stacks and queues are pretty vital parts of quite a bit of the programming I've done and Mr. Harold's ignorance about the subject leads me to believe that he must have slept through that part of the course, since I assume that he does have training in the field of Computer Science. Hell, I learned that stuff ages ago, even though I went to college in the early '70s when that type of course didn't exist at the University of Michigan.
Cees deGroot has a similar viewpoint here.
(Trackback)
17 October 2005
Ruby O-O programming and the big time
The local open source has scheduled for the next monthly meeting a talk on Ruby. The description says in part, "ruby is a newer, popular interpreted language supported on many platforms. It is respected for its powerful object oriented design."
Count me unimpressed. I guess if you compare it to perl, it does have a better implemented object facility than perl does. I'll take a less "fashionable" GNU Smalltalk over Ruby any day. GST is meant to be a scripting language just as Ruby is, however GST uses a different syntax that isn't borrowed from procedural based languages like C, C++, and Java (Yeah, I know C++ is supposed to be an OO language).
Hmmm, maybe I'll do a talk on GST at one of the upcoming meetings.
Count me unimpressed. I guess if you compare it to perl, it does have a better implemented object facility than perl does. I'll take a less "fashionable" GNU Smalltalk over Ruby any day. GST is meant to be a scripting language just as Ruby is, however GST uses a different syntax that isn't borrowed from procedural based languages like C, C++, and Java (Yeah, I know C++ is supposed to be an OO language).
Hmmm, maybe I'll do a talk on GST at one of the upcoming meetings.
02 October 2005
More on OPML
...
OPML spec, but given that the spec is in a place that doesn't appear to be any place "official", I don't know how useful it is.
Maybe this is as professional as Dave gets. I'm not saying that Mr. Winer is a hack, in fact I think he's a bit of a visionary in the internet space, but it would be nice to see him link up with people or organizations that will fill in the blank spots that Dave leaves in his work. He's more of a hacker who needs some documenters to make his stuff into something usable to everyone else.
I'm comparing this work with the RDF Site Summary (RSS) 1.0 docs, which includes a more complete spec and an RDF definition. Another usable spec is the RSS 0.91 Spec, revision 3. In contrast to these specs is the RSS 2.0 Specification, that's another example of Dave's work. I'm not aware of sample files for the OPML spec like there are for the RSS 2.0 spec.
Subscribe to:
Posts (Atom)