| The Ultimate Business Machine Technology, business And, not least, about by: Bernard Teo
|
Thu 18 Aug 2005 The Chocolate FactoryCategory : Commentary/wonka.txt Well, what do you get when you cross Steve Jobs with Michael Jackson? Willy Wonka, as played by Johnny Depp. Tue 16 Aug 2005 Drag and DropCategory : Technology/draganddrop.txt I've released new versions of DNS Enabler (1.1.6) and WebMon (1.1.5) with drag and drop support. This will allow a DNS Enabler user, for example, to re-arrange the DNS entries in any order that he wants. I've also figured out how to add a toolbar to our applications. And to start working with icons. In a couple of weeks, I think we would be ready to release Luca with an embedded SQLite database so that it'll be really easy to set up and run an accounting system. Hopefully we can evolve it into something more like a financial management tool. And I'm working on a new version of Postfix Enabler (which won't be called Postfix Enabler). So things are moving. The whole point of going through this litany of works-in-progress is to reflect on how very productive the Cocoa development environment is. If you've gone through the dark ages when one vendor after another abandoned the Mac platform, you may have a sense of how liberating all this feels to me. I used to work with an EIS/data mining tool called GQL and also something called Data Prism. They were a revelation because they were so much cheaper than the incumbent host-based systems. Data Prism and GQL cost hundreds or, at most, thousands of dollars but not the hundreds of thousands that the IT industry, then, assumed that decision-support tools have to be. And they ran on the Mac. At first, only on the Mac. And they were a revelation because they showed how a GUI could help you visualise the relationships among the data. Remember then that the host-based systems (by that I mean those running on DEC and HP minicomputers, etc) were all character-based. So, for a while, people saw the Mac in a different light. Then we went into a tailspin (Sculley, Spindler, Amelio) when one product after another dropped support for the Mac and became Windows-only. When GQL became Hummingbird, that was the end of the Mac version. Nowadays, I feel we can build all these ourselves. The drag and drop needed to implement a data mining GUI - one that will allow you to pull the pieces together to construct a database query - that's there. The access to a wide range of databases - Oracle, MySQL, and SQLite. That's also there. The ability to work with tables, sorting them, rearranging column order, hiding and showing columns on the fly. These are also there in the Cocoa toolset. The ability to simply drag and drop the result into a spreadsheet, a presentation tool, or a publication tool - they're already there. It's a waste to use all these capabilities just to build graphic tools or Final Cut Pro, etc. There's a wealth of business tools waiting to be built. So we can drag and drop them into our businesses. Thu 11 Aug 2005 WebMon 1.1.4Category : Technology/WebMon114.txt This is WebMon 1.1.4.
I've re-arranged the workflow so that WebMon works like Postfix Enabler and DNS Enabler, at least initially, where you work locally on the server. It takes only a couple of minutes to turn on PHP, WebDAV, and SSL on the server. And to check that WebMon's web server monitor feature works. But unlike the current versions of Postfix Enabler and DNS Enabler, WebMon works across a network. This is potentially very powerful but it may not work for everyone because many things could go wrong while setting up the remote connection. Re-arranging the work flow this way allows me to unveil the remote admin capability in a couple of steps, and hopefully I can take the user across safely, while offering him or her useful things that work at each step of the way. It may need a little bit more work and the interface needs to be a little bit less harsh but I think I'm getting there. A lot of Cocoa's appeal as a development platform is the way you can use Interface Builder like a storyboard without worrying about program code, i.e., you can keep on arranging and re-arranging the pieces of the interface and concentrate on telling the story without the technology getting in the way. In other words, you can be the artist when you want to be and the geek at other times, depending on the task at hand. Art and Technology - a new unity. Now where have we heard that before? Right, the Bauhaus. Sat 06 Aug 2005 WebMon 1.1.3 ReleasedCategory : Technology/WebMon113.txt Of the three - Postfix Enabler, DNS Enabler and WebMon - WebMon is probably the most buggy of the lot. That's because it's the first one where I'm building in remote configuration, right from the start. I could make WebMon work in standalone mode, where you do the configuration locally on the server (for WebDAV, PHP, SSL, etc) and that would improve its stability no end. But when it comes to monitoring the web server activity, it's really more convenient to do it remotely. Who wants to walk to the server just to see what's going on in the web server log? So, what I've done in WebMon 1.1.3 is to add the ability to work directly on the server, while keeping the option to work remotely. A good strategy to use with WebMon is to start working at the server. In the WebMon setup panel (accessed from either the Help or Preferences menu), you enter the server name as "localhost", first line below :
And then you turn on WebDAV, PHP or SSL. And test that all these things work, including the ability to publish iCal calendars in the WebDAV folder (which you could also use as a backup store, like .Mac's iDisk folder). If these are OK, you could start preparing to use WebMon for what it was originally intended to do - as a Web Server Log Monitor. You could enter a line like the second line in the screen shot above, where you call the Web Server, userName@localhost, where userName has administrative rights to the machine. The "@localhost" notation signals that, although you're working locally at the server, you will be "pretending" to connect remotely. WebMon will try to set up a remote login connection while you're still working locally at the server machine. This is useful because you can ignore issues like network speed and timeouts, for now, while testing conceptually that your machine has been set up properly to support remote login connections through SSH. If you're using an unmodified, garden-variety OS X machine, you should have no problems getting this set up through WebMon. But some systems have custom modifications that WebMon can't find its way around. If Step 2 is OK, you can now run WebMon from another machine and get it to connect to your server, using the userName@domainName notation, like in line 3 in the screen shot above. If you get to Step 3, and you can find a reasonably fast connection to your server over the network, you should be able to monitor the server's log file and change your server's configuration, all from a remote machine. And the whole idea with WebMon is that you can monitor and configure any number of server machines from this single control machine. If Step 3 doesn't work, because of connection issues, etc., at least WebMon would have been useful in Step 1, where you limit yourself to doing things locally on the server machine. I hope WebMon works for more people now. Thu 04 Aug 2005 DNS Enabler 1.1.5 and other thingsCategory : Technology/DNSEnabler115Released.txt DNS Enabler 1.1.5 is out. I've also added a new item in the Help menu to allow the user to de-install DNS Enabler completely, stop the name server, and restore eveything that DNS Enabler touched to its pristine state. I've put an expiry date to this version because I do intend to sell it as a commercially supported product - when I'm sure that it's totally bug-free and really useful. But it's going to be priced like Postfix Enabler. Hopefully it will be deemed low enough to escape the radar of the pirates but I doubt it. But pricing Postfix Enabler super low at 9.99 didn't save me from being called "stupid and greedy" at MacUpdate. Makes you wonder about the phrase "nothing more to lose". But let me tell you about the book I've just read and enjoyed. It's "Juice - The Creative Fuel that Drives World Class Inventors" by Evan I. Schwartz. That's the third (okay somewhat) good book I've read from him (the others being Webonomics and Digital Darwinism). But there's this recurring theme in Juice that says that, with every new invention or product, you've got to work out who wins and who loses with the introduction of the product. Because the ones who think they're going to lose are going to counter-attack. So you see these comments at MacUpdate and Version Tracker? They may not be just from the users. Mon 01 Aug 2005 DNS Enabler 1.1.5Category : Technology/DNSEnabler115.txt This is how DNS Enabler is shaping up. A couple of more things - get it to handle CIDR (Classless Interdomain Routing) formatting (I lost it in the re-write) and make it more forgiving of user errors - and I should be able to release it. I hope it would be soon because I found a bug in 1.1.4 (when it's serving more than one domain and the user re-sorts the table).
This shows DNS Enabler handling the data entry of a domain described in Chapter 4 of the Paul Albitz/Cricket Liu book, DNS and BIND. I started with a very simple design to get DNS running on a local network with the minimum of data entry. Then I got a bit more ambitious (maybe I got carried away), while trying to hang on to the same basic design. I've tried to cut away as many things as I could but I don't think I can prune any more. Everything that is there is there for a reason. DNS Enabler is now able to handle multiple virtual domains, local as well as public networks (I hope to prove this soon), multiple sub-nets, aliases, MX records (including specifying back-up servers on other networks), multi-homing (one IP address shared by many hosts from different domains or each host on its own IP address or one host name spread across many IP addresses), and creating the right number of reverse pointer and CNAME records. And it does it all on one window, across just three columns, which can now be freely sorted - with the primary host name on the left, its IP address in the middle, and its function on the right (e.g., whether it is also known by another name, or whether it acts as the domain's mail server, and whether it has an associated back-up mail server, etc.) I hope by tomorrow evening this will be out. Then I can go back to doing a standalone version of WebMon that will work locally on the server machine (and that doesn't depend on a remote SSH connection being set up, which is a needless complication if you just want to set up WebDAV, PHP, and SSL). Thu 28 Jul 2005 DNS Enabler 1.1.4Category : Technology/DNSEnabler114.txt I've updated DNS Enabler to include the ability to update the domain's SOA (Start of Authority) information including the zone's default TTL duration (Time to Live). The serial number for the zone will also work properly now, and it increments with each modification to the zone files.
This all in preparation to make DNS Enabler work on a public network, so that it'll be able to transfer data to a secondary name server when the zone data changes.
It's on to MX records next and, after that, I'll revise all the documentation. Like Postfix Enabler, WebMon, and all that, this is the interface I've wanted to build ever since I figured out how to do things manually. In fact, DNS was the first thing that got me on this track. I was using AppleShare IP in the pre-OS X days to serve out our Cutedge web site. I started by using MacDNS, which was buggy, and then Men and Mice's QuickDNS Pro. On moving to OS X and OS X Server, I found that it didn't have a GUI for configuring DNS and QuickDNS Pro didn't run, then, on OS X. I was on a leased line public network then, and so I had to learn how to configure DNS by hand. Surprisingly, I managed to do so, which piqued my curiosity, rekindled my interest in Unix, and I wondered if I could do the same on a plain OS X machine. When I found that I could, it led to the idea that maybe a mail server will run, too, on that plain OS X machine, i.e., on my garden-variety iBook. And that has led to all the stuff you see on this page. So, DNS marked the start of this journey and that's why getting DNS Enabler to this stage has been of some significance to me. And I wanted to do something which didn't copy the QuickDNS Pro interface, but to try to improve on it in terms of its usability by someone with very little DNS knowledge beyond the essential basics. Finding a way to allow the user to specify an MX record within the confines of the DNS Enabler interface, as I have designed it, is the next challenge. I think I have an idea. By the way, my iBook is now more than three years old and it's falling apart. It's trackpad has become unusable. But then, this is my most useful, most used, most productive computing machine ever. I will always love this iBook but I'm looking forward to replacing it with the new 12 inch iBook that's just come out. That is the iBook I've been waiting for. Yes, spend wisely. Right on time. Tue 26 Jul 2005 Sendmail Enabler's New HomeCategory : Technology/SendmailEnablerHome.txt It seems like there's still quite a lot of people using Sendmail Enabler. Remember Jaguar? That's a long time ago. I'm moving things around and Sendmail Enabler gets a new home to see out its days. Sendmail Enabler for Jaguar can now be found at : http://www.cutedgesystems.com/software/SendmailEnabler/index.html. Sun 24 Jul 2005 AuthenticityCategory : Commentary/authenticity.txt I remember why I brought up the book, "Trading Up", a couple of posts ago. I enjoyed that book (e.g., the story about how Victoria's Secrets came to be) and I'm probably not doing it enough justice, but the gist is, you either position your products at the low end when you can slog it out on price, or at the high end where you've developed an unassailable brand. But the worst is to be caught in the middle where you'll be decimated. The best is to have what is called a "mass-tige" product, that is, a prestige product that confers status, which costs somewhat more than a standard product, but not too high that it's out of reach, yet will return significantly higher returns, even as you sell a whole lot of them through your ability to make them in consistently good quality. This turns the traditional supply/demand curve on its head because when you're in the happy position of owning a masstige product, you can sell it at a high price and find that instead of selling less of them, you can actually sell even more of them because the price reinforces the perception of quality. What, then, are the characteristics of such "masstige" items, so we'll know one when we see one, preferably when it's within our power to bring one into being? That was the purpose of the book to describe. But among the characteristics are the following, if I remember them off the top of my head: they've firstly got to be clearly better built - better made, better designed, with a thoughtfulness that comes from an attention to detail - and they've got to be consistently good, even where they have idiosyncrasies and variations. They've got to include a fair measure of technology, otherwise you wouldn't be able to reconcile the conflicting objectives of achieving consistently high quality while making production in quantity. Yet the technology is tempered by the existence of a story, to wrap the product in meaning, so that it speaks with a voice or an authenticity that will connect with the heart of the user. There are other characteristics but these are the major ones. And they're the ones that have a connection with why I felt that the Mac had been the right business machine all these years. It's the idea of authenticity. It's the idea that we have thought through all the reasons why we're in business, and that we've taken the utmost care to breathe life into our products, and that we will not stop at anything but the best to deliver that experience to the user - including harnessing the best supporting information systems so that it radiates quality and life and thoughtfulness and craftsmanship at every step of the way. Of course most of those businesses covered in the Trading Up book probably aren't using the Mac but my point is, if you really know, would there be any question what you would choose? For me, the decision is obvious. I can't achieve the authenticity, so to speak, if I didn't choose the Mac. It's not like I'm a crazed Mac freak, but it's clearly the only one that has a chance to resonate with all the other good intentions, if all I have is a choice of Windows, Linux, Unix, or the Mac. I mean, how can you speak with a singular voice, if you choose to use something simply because everyone else is using it? It's like, while reflecting on the potential of the technologies to properly support the business, have you asked "what does the brick want to be"? Have you listened to that inner voice and see how you can harmonise every aspect of the business so that it speaks with a consistent tone, colour, grace, and optimism? It's hard to see how the answer could lie in something as dead and soul-less as Windows. That's why it's so hard to do the truly authentic thing. It's running against the crowd and listening to that inner voice. And keeping the faith. Airport Extreme Base Station Firmware UpdateCategory : Technology/AEBS551.txt There's a problem with the Airport Extreme Base Station Firmware Update 5.6 that prevents the base station from connecting to the ISP over PPoE. It's not consistent. An AEBS updated to 5.6 continues to work with PacNet but fails with Singnet. I managed to find the previous 5.5.1 update at the Apple site. After loading that in, the Base Station works again. For a while I thought I had lost a base station. Avoid the 5.6 update if you can. (Actually, avoid it like the plague.) Sat 23 Jul 2005 To The LibraryCategory : Commentary/library.txt Where I really want to go today is to the library. A few years ago, they decided to tear down our historic, original, very first National Library, amid much wringing of hands and gnashing of teeth. In order to provide the new in-city Singapore Management University with contiguous access to the Singapore History Museum so that they make one lush campus (will it ever be?), a roadway has to be diverted and the Library had to be sacrificed. Whatever the merits of the decision, a new high-tech replacement was opened yesterday. And since the Library is the institution I cherish the most as a citizen of this country, this is where I want to go today - to re-establish my connection with the Place, shifted though it has been a couple of roads away. I need to return a book that I had finished, "Trading Up", so that I can borrow "The Company of the Future". The last I check, that's still there. Just a word about "Trading Up" before I return it. I've learnt that it's important to try to understand what makes people value what they value and are prepared to pay for. I've sat in meetings and faced countless inquisitions about why I favour wasting tax payers' money on Apple's high priced gear, why not the cheapest possible PCs, and after coming out from the battering, I see these same guys driving away in their red Honda Preludes, BMW, whatever, and I want to ask, how much is your mind worth? And as an IT sub-contractor, don't you know that a customer would rather pay for his Merce, than spend anything more than peanuts on the IT system? The more the Mercedes costs, the better he feels, because it shows he's rich. The less he pays for an IT system the better, because it shows that he's smart about controlling his operating costs. Between these two forces, a guy trying to make a living as a software/systems developer dies a thousand deaths. For years, Apple has suffered through these with the Macintosh. Because we don't value the mind more than we care about showing our status, Macs have been consigned to the ghetto while BMW sales rise, Steve Jobs' fruitless attempts to compare Apple with BMW be damned. Until the iPod. Which is Apple's sweetest revenge. So it's important to understand the psychology of buying - if only to know when to get out of the way of the tsunami of people's self interests. I say this with a lot of feeling for the idealists. Even if you don't want to go with the flow, you should know enough to get out of the way. And live to fight another day. And perhaps, like Apple, learn to turn the force your way? Where do you want to go today?Category : Commentary/wheretogo.txt Okay, I'll borrow MS's slogan for Windows, and back up a little to reflect on where I'm trying to go with WebMon, DNS Enabler, and Postfix Enabler. I've started building these things, initially, for myself. The web, mail, DNS, calendaring system, accounting system, integrated business database system, payments system - overlaid with security and encryption - these are what I think of as the life-sustaining elements of any business. Of course, you've still got to go out and sell your wares. But these are the stuff you'll need to back you up, and they, in turn, need to work flawlessly and easily if you're going to channel all your energy towards making the products and services, and bringing in the revenue. And none towards feeding the system. I'm making all these so that I can turn them all up quickly if I need to change a server, move to another home, another country, anywhere where they have an Internet connection. I've got the web, the mail, calendaring and file sharing over WebDAV, DNS, and SSL done (or at least the basics). Coming on is the accounting system, and a way to snap on a PayPal payments interface, and link it to the accounting system. And tie in this no-frills blogging system with perhaps a GUI editor. Currently Postfix Enabler and DNS Enabler lack the ability to help you migrate the data over from an older machine or system. I had thought that I've already eliminated a lot of things a person has remember. But I understand now that people would need a way to restore their previous configuration with just one click. After all, that was what I said was my stated goal - to get back on-line fast. I'm not there yet. It needs more work. And some people have problems with PayPal. But believe me, I couldn't do this without PayPal - not if I'd like to get paid for sharing the tools and providing the support. I hope PayPal gets better because I don't have a better or cheaper alternative now, and I think they will get better. So this is where I want to go. I hope that others will also find the tools useful and agree they're worth paying for, and that they could build their businesses on the Mac, which would be nice bonus because two years ago when I took this route it was really to find a non-IT business we could run, but with all our own tools, because Hai Hwee, my wife, and me got tired of working with PC-myopic IT managers, overseers/information architects, Windows, and all that corporate IT scene, and I wanted to be near our kid, and be a daddy rather than a wallet (as someone puts it). And we're still looking for it - a business we could run like a high-performance machine - so that we, and everybody else working in it, could live our lives like human beings. But watching the payments coming in for Postfix Enabler has given us a glimpse that maybe, if anything, the underlying systems do actually work. And they can scale. And they're all running beautifully on the Mac. As I've argued they would. And so, on with the search, to build a real business on it. Read more ... |
Mac@Work
Services running on this server, a Mac Mini running Mac OS X 10.9.2 Mavericks:
all set up using MailServe, WebMon, DNS Enabler, DNS Agent, VPN Enabler, Liya and our SQL installers, all on Mavericks. |
||||||||||||||||||||||||||
