danbricklin.com/log

Starting November 18, 2012
An overview of HTML 5, PhoneGap, and mobile apps, More JavaScript news, An old video about the history of pen-based computing, Links for Xconomy Mobile Madness panel, Some things happening with JavaScript, 10 days with a Surface Pro, MS Surface Pro: It's the spreadsheet, stupid, 30 years since Lotus 1-2-3, The Developer's Challenge in 2013, End of a podcasting era but NYTimes #56,000, Original VisiCalc running in a JavaScript IBM PC emulator
18Nov12
2012_11_18.htm
Thursday, May 2, 2013 
An overview of HTML 5, PhoneGap, and mobile apps [link]
In my discussions with business IT people, I'm finding that the use of HTML5 and perhaps "hybrid" architecture is becoming more and more commonly accepted. However, I'm also finding out that most people don't understand exactly what this means, nor have a clear picture that lets them differentiate between variations of that architecture supported by the multitude of development environments. I wrote an essay to fill in a lot of the details to help people understand that picture.

Read "An Overview of HTML 5, PhoneGap, and Mobile Apps" in the Writings section of my website.

Sunday, March 31, 2013 
More JavaScript news [link]
Two additional things on the JavaScript front that I noticed. First, Dave Winer, the innovation-filled pioneer who made a popular outlining application in the early personal computing days, then helped teach many of us what blogging was all about while building and evangelizing tools and standards (including RSS), and pioneered the concept of podcasting, just came out with a JavaScript-based product. It's a simple outliner, called "Little Outliner", that runs completely in a browser, including using browser-based storage. It's the first offering from his new company, Small Picture.

Second, Mozilla and Epic Games announced that they were bringing Epic's Unreal Engine 3 (software for 3D games) to the browser. This uses special JavaScript speed-up techniques (asm.js) and WebGL (graphics hardware support). The demos are pretty impressive. The speed requirements for most regular applications are minimal in comparison to the 3D-fly throughs they are showing. Just the idea of this is pretty amazing.
 
An old video about the history of pen-based computing [link]
I've been looking at some of the old videos I have, and, as you may have seen on this blog, I've been posting some of the general interest ones on YouTube.

Back in March of 1992, Jean Renard Ward, a software developer then working with me at Slate Corporation, gave a presentation at the Boston Computer Society about the history of pen-based computing. He gave the talk a second time in our office in front of a video camera (though he still used an overhead projector for the visuals). I received his permission to post it publicly, and tweaked it a bit today (the original audio was quite low, so I added gain) and posted it to YouTube. It's a fascinating view of the many, many devices produced before that date (going back at least to 1914).





Tuesday, March 19, 2013 
Links for Xconomy Mobile Madness panel [link]
I'm on a panel later today at Xconomy's Mobile Madness 2013 conference at Microsoft NERD in Cambridge, MA. I have to look into the future while representing the past, I think. There are some "old" essays that I wrote that I might make reference to, so here are the links. (They are worth reading in any case.)

"What will people pay for?" (written July 2000) -- the value of social connections.

"The Cornucopia of the Commons: How to get volunteer labor" (August 2000) -- Build systems where increasing its value is a natural by-product of people using the system for their own benefit.

"When the long tail wags the dog" (written February 2006) -- General purpose systems win. People figure out how to apply what they know how to use to what they need or want to do.

Also, you can follow me on Twitter: @DanB.

Tuesday, February 26, 2013 
Some things happening with JavaScript [link]
A lunch that I had with an old colleague and some announcements from Mozilla and Microsoft led me to write an essay about the ongoing evolution of JavaScript (and the environment in which it is run) into a dominant platform for user interface programming.

Read "JavaScript rolls on (as does SocialCalc)" in the Writings section of my website.

Wednesday, February 20, 2013 
10 days with a Surface Pro [link]
I've been using a Microsoft Surface Pro 128GB for about 10 days. Thanks to perseverance, luck, and the very great staff at the Microsoft Store on Boylston Street in Boston, I was able to get one when it first came out. (The store manager let some of his staff stay in a local hotel so they could be open despite the shutdown of Massachusetts public transportation and private driving due to 2+ feet of snow, and he set aside the units many had "reserved" by coming in previously and getting a "reservation card". Bob Frankston's son happened to be stuck that day within walking distance of the store and picked them up for us. I got them the next day when driving resumed.)

A lot has been written about the Surface Pro, so I won't say too much here. What I found was that it was just a curiosity at first, since it was pretty much like my Surface RT, except heavier and with a higher-resolution screen. That changed over time. I spent some time on and off over the next few days loading it up with most of what I run on my main Windows machine. I have Thunderbird, with all my mail accounts and addresses, I have Microsoft Office, Access, Adobe Cloud (Photoshop, and lots more), photos, Firefox, Chrome, my speaking material and the apps I use to show it, and Microsoft SkyDrive. It's now a "real" machine. It's not like my old netbook, where it seemed like a "real" Windows computer but way slower and with a screen with too little resolution. This is a fast enough machine, with enough memory and a high enough resolution screen for real work. I am quite used to my 11" MacBook Air (that is, using a small screen and device), and this feels enough like a Windows equivalent. Except...it's also a tablet. It does especially great as an "on table" tablet when reading during lunch. But the touch interface really does add a lot to using apps that are tuned to it, and even those that aren't. I really love the UI of the "Modern" style Internet Explorer. IE gives you the whole screen, but gets you the tabs and other "chrome" with a simple gesture or two. Unlike on other devices I've used, you don't have to carefully tap small icons to get favorites or tabs. You just do a swipe in from the right place or a tap on a large area. These "lower precision motions" are really nice for common operations.

I had the experience at one meeting of suddenly having to go to my email and retrieve a Word file with a document we were discussing. I used the Surface Pro's pen to make notes on the document in Word. Then, at the end of the meeting, another participant asked me to email it to him, and he was surprised that he could read my handwritten notes directly in Word on his Mac. He was taken aback by that and now saw the value of this device. I can see the same thing happening in meetings all the time when people see a Surface Pro in action.

It's nice to have an accessible file system. No... it's really nice to have an accessible file system.

If you're comfortable with Windows, and like the idea of a device that you can make do a lot of things with software that already exists and that you probably already know how to use, then the Surface Pro may be a useful compromise for you.

Wednesday, February 6, 2013 
MS Surface Pro: It's the spreadsheet, stupid [link]
Today a lot of "official" press reviews of the upcoming Microsoft Surface Pro combo laptop/tablet are showing up. I have yet to get mine (I guess this blog doesn't count as a major news outlet) but I have one that I "reserved" for me at a local Microsoft Store (I had to do that in person, like everyone else) which I hope to pick up this weekend.

I received an email from a user of my Note Taker HD iPad application. A happy, long-time user of my app, he's switching to a Surface Pro this week. Why? He told me that he finds challenges in using the iPad for business, especially with spreadsheets. That got me thinking.

Seven years ago in 2006, I wrote an essay titled "When The Long Tail Wags the Dog". Here are two relevant parts: First, I showed why general purpose devices are desirable and preferred by users. Apple exploited that desirability by getting people to think that whatever they needed, they'd find "an app for that" on iOS. In the world of business right now, you know that an awful lot of important business applications, especially custom internally-focussed ones, are available for Windows machines, often only Windows machines. If there's a business app you'll need in the near future, it's probably available for Windows so you know a Windows device will serve that not-yet encountered need.

The second point in my essay came from an interview I did with Donna Dubinsky, ex-Palm Computing CEO. In addition to using one or more of the few "popular" built-in applications (calendar, address/phone book, etc.), "...rather than adding 3-5 applications [to those existing built-in ones] on a device, customers would find one that was compelling for them, and that to them was a make it or break it thing. It might have been the stars and where the stars are, it might have been world traveler applications, or it might have been querying a detailed database at work, but there was always some additional compelling application for the Palm owner."

What this user pointed out to me is that there is something unusual on the Microsoft Surface Pro: The "one" application that many people need is Microsoft Excel. Not "Excel compatible" or "Excel minus x, y, and z" (like on MS Surface RT), but full Excel. Full Excel is an awesome product, as in awe-inspiring in its breadth and depth. Whole $100M++ companies have been built around selling data analysis systems that are intimately tied into Excel. Others take advantage of its ability to work with thousands and thousands of rows of data and its powerful programming language.

Having done some work a few years ago looking at making my SocialCalc spreadsheet more "touch friendly", I saw that there are lots of challenges to deal with the coarse pointing accuracy of normal touch screens. Excel is tuned to, and highly exploits, a mouse pointer with hover (pointing without touching). An acceptable alternative, I found years ago, is a high-resolution pen, like that used in the old Tablet PC's. Excel's spreadsheet displays pack a lot of data on to the screen and need a high resolution display and a high resolution pointing device. The Surface Pro has such a pen, and also has a touchpad on the keyboard and supports a wireless or wired mouse in all applications without modification. iPads do not have such system-wide pointing devices nor support for hover. There are add-on special purpose pens from 3rd parties, but they require application support, which is currently very limited and needs to be done for each manufacturer, and cost an additional $80+.

In my case, one of the reason's I'm buying a Surface Pro is to run a Windows-only program that I need. While I'm at it, I'll make it my second machine for my Adobe Cloud subscription for Photoshop, Audition, Premiere, etc., when needed. With a Microsoft Office subscription, I can also add the Surface for the Powerpoint, etc., stuff I need and still run some other old Windows software I use when doing presentations which serve me better than the iPad. (I tried setting up an iPad for my talks, but with the different parts that have evolved over the years, it was too much work. I still use a netbook or a MacBook Air with some 3rd party software when I give presentations. Sometimes using a file system is the easiest thing if you've spent 50 years thinking that way.)

As I pointed out in my recent "The Developer's Challenge in 2013" essay, a combo device like the Surface Pro lets you treat it as a tablet when you want, or like a laptop. You don't need to switch devices, you just switch how you treat it and choose applications accordingly. If you really need some particular iOS program, an iPad mini or iPod touch are fine, inexpensive devices and weigh very little to run those new "must have" apps, but in all likelihood, you are probably already carrying an iPhone which serves that purpose. For the common tablet use of big-screen reading, web surfing, and watching video, the Surface, like most tablets, fills the bill.

So, perhaps the "killer app" that will drive Surface Pro sales is the "original" killer app of the personal computing world, the spreadsheet. We'll see.

Monday, January 28, 2013 
30 years since Lotus 1-2-3 [link]
Over the weekend, I started seeing a variety of tweets and Google Alerts about an important milestone: January 26, 2013 is the 30th anniversary of the release of Lotus 1-2-3. The head of the project, Mitch Kapor, was interviewed by The Register.

There is much to learn from 1-2-3. Many newcomers to the world of using personal computers in business, unfortunately, don't know much about it. While VisiCalc gets to be in the history books for being the first of its genre, only the currently most popular are being taught (Microsoft Excel and Google Docs, I assume). (See my blog post from February 5, 2001, about VisiCalc being an easy question on the TV game show Jeopardy. See "VisiCalc" on my main web site for more about the program.)

Geoffrey Moore wrote about the "Tornado" -- the time period where the mainstream market adopts new stuff. Lotus 1-2-3 is the poster child for the winner at this time for the initial adoption of personal computers in business. It both rode the tidal wave and helped push it along.

When Mitch first showed me 1-2-3 privately before it was announced, I was blown away. Wow, what a great product! I thought to myself: "This will sell $60 million the first year." Lotus was apparently only predicting $5 million, but the actual ended up a bit over $50 million -- on par with Microsoft. That was just the start of its growth.

There were many spreadsheet programs on the market when Lotus 1-2-3 came out, including VisiCalc, SuperCalc, Microsoft's Multiplan, and Context MBA. Mitch compared 1-2-3 in the demo he showed me to Context MBA. Context MBA was programmed in a slow, high-level, byte-coded language, as I recall, while 1-2-3 was in assembler for the IBM PC. He told me that Lotus tried to meet similar goals to what we had originally targeted with VisiCalc, such as keeping up with fast horizontal and vertical scrolling. It felt at least as good with big sheets and the full screen of the IBM PC as the original VisiCalc felt on small sheets on an Apple II with much fewer characters on the screen. Context MBA felt like molasses in comparison. Both had lots of features (which is why he compared the two instead of using the lesser-featured MultiPlan or SuperCalc).

One major thing about 1-2-3 was that the implementation and design choices were very pragmatic. It used what was good from before, such as lots of features from VisiCalc, but changed things for the better where it was most helpful. For example, it had VisiCalc's  "A1, B1, C1" row/column naming, unlike the techie "R1C1" and the like of Multiplan, but it added optional "$" characters in the cell coordinates instead of the manual "relative or absolute" of VisiCalc. It took advantage of the new hardware, making great use of the IBM PC's keyboard, even coming with a molded template to label the function keys. It used the extra memory and the knowledge that there could be two disks to provide much better command prompts and a real context sensitive help system. It addressed the core complaints about VisiCalc (lack of thousands separators in numbers and lack of individually settable column widths) and much more. It was written in tight assembly code, accepting the limitation of making porting to other computers more difficult (it couldn't run on Apple's 6502-based computers or many others) to get the speed needed on the soon-to-be-dominant IBM PC (which it helped make dominant). (Context MBA and Multiplan were not written in assembly language for one machine -- they followed the "write once run everywhere" higher-level language technique that seemed appropriate when you wanted to provide products for the myriad different computers of the day.) 1-2-3 included built-in graphing and simple list management -- two major uses of spreadsheet data.

While VisiCalc concentrated on just being able to do spreadsheets at all, Lotus 1-2-3 went to the next level and addressed the final printed output much better, with more number formats, variable column widths, long labels, and that very-hard-to-do-by-hand graphing. And it did all of this with greater speed than anything else. Speed and fluid operations matters, as the Palm Pilot later showed (with its instant page turning in response to taps), and then the Apple iOS products showed after that.

The code itself stood the test of time and for years beat out most other products running on the hardware for which it was designed. It wasn't until a platform switch occurred (GUI) that the torch was passed to the next dominant spreadsheet, Excel.

The team Mitch assembled to get the "full" product (including documentation, sales, etc.) out the door had some of the best positioned and most experienced (with personal computing) people in the world. He also worked with one of the most tied-in to the personal computing business venture capitalist, Ben Rosen. A dream team at the time and in hindsight.

To help give you some feel for the excitement after the launch, I have two videos that you might want to look at. The first is a promotional video that Lotus showed to dealers at conferences (and shown a bit in Part 1 of the Fall Comdex 1983 video that I posted last year). Produced just two years after MTV started, it's in the style of a music video. It shows how excited an office gets when they can move from the "floppy shuffle" of the old products (like, clearly, VisiCalc and its co-products) to the integrated spreadsheet, graphing, and "database" of 1-2-3.

The video is "Lotus 1-2-3 Rocks" and is embedded here:



The second video is one shot by Adam Jackson showing Mitch Kapor presenting at Startup School 2009. He shares another Lotus promotional video that covers the first 5 years of the company -- watch especially the beginning and compare it to today's Facebook, etc. Lotus, founded in 1982, went public on October 6, 1983, less than a year after shipping 1-2-3. That made the evening national news, as you'll see. Watch Mitch at Startup School 2009.

Getting personal computers onto the desks of office workers everywhere was a very important step in the history of computing. Lotus was a major factor in taking this step. Their later product, Notes, I believe helped get "wired" computers onto those desks and hastened the adoption of web browsers for many reasons (it's a lot easier to get people to try new software and services when they already have the expensive hardware and it's wired and ready to go). While the old Lotus is not around in its old form, its employees have gone on to help create other great things in the computer industry. Mitch has continued his role as an industry statesman, and I hope is enjoying this anniversary.

People frequently ask me how I feel about 1-2-3 overtaking my product VisiCalc. While it always feels bad to lose your position as a leader, and not get to participate as much in the benefits that come with that position, I'm really happy that at least it was 1-2-3 that took the mantle from VisiCalc. Mitch and Jonathan Sachs were our friends and they made their product a follow-on (it could read VisiCalc files, so you could move your spreadsheets from VisiCalc to 1-2-3 to Excel to Google Docs without retyping) keeping a lot of the "DNA" of our ideas. Lotus improved on the design of the electronic spreadsheet, so it stayed a major productivity tool. Mitch kept his company here in Massachusetts (Mitch had moved back from Silicon Valley to found it). And our product is still the first in the line and is not forgotten. As a child of the 1950's and 1960's, to know that you made something that changed the world, and that it lives on in products that acknowledge your starting point, is something most people could only dream about and for which I will be forever grateful. The launch of Lotus 1-2-3 helped make that happen and brought personal computing to a large part of business in the process. Happy 30th!

Wednesday, January 16, 2013 
The Developer's Challenge in 2013 [link]
In November 2010, I wrote an essay titled "The Developer's Challenge in 2011." It covered many of the challenges to the developer, especially the business application developer with server-based backends, stemming from the coming addition of the Apple iPad to the "must support" devices that included smartphones and desktop computers. Now, a little over 2 years later, that challenge remains and most of the forecasts have come true.

The hardware landscape has continued to evolve, and additional new devices have entered the arena. I wrote a new essay that covers some of those newer devices.

The previously listed devices that need to be considered were: Smartphones, tablets like the iPad, laptops and desktops, and large screens shared in a room (such as projected).

The new devices I now want to include as separate are: The mid-size tablet (for example, the iPad mini and the Google Nexus 7), the tablet/PC combination (most notably the upcoming Microsoft Surface Pro), and the TV-bridge device (such as the current Apple TV device). These may not at first blush be considered different, but I believe that they are.

Read "The Developer's Challenge in 2013" in the Writings section of my website.

Saturday, December 29, 2012 
End of a podcasting era but NYTimes #56,000 [link]
While exercising today I listened to the last IT Conversations podcast, "All's Well That Ends Well". It featured Phil Windley interviewing IT Conversations founder Doug Kaye earlier in the month.

Doug created the second podcast ever (the first was one recorded by Christopher Lydon posted by Dave Winer, if you use Doug's definition of podcasting starting with using the RSS enclosure tag). ITConversations.com started in 2003. The ongoing series of podcasts of many conferences, interviews, and "talk" shows like the Gillmor Gang, showed many of us the value of podcasting and also was of great educational value. Doug was my main mentor as I learned how to podcast, as he was for many others.

Podcasting is now well established, and from what I hear from my friends and others, it is a major way that lots of NPR and other programs are experienced. As Doug explains, it was time to end ITConversations -- it is not needed in the way it was. It is a pioneer that helped change how we get information, but that way is now the mainstream. They have found long-term repositories for their previously recorded programs (e.g., the Internet Archive for techie ones).

To me, this was kind of sad. The end of an era.

Coincidentally, when I finished listening to that recording, I then listened to some of Marco Arment and Dan Benjamin's "Build and Analyze" podcast, episode #108 -- the last one, as Marco decided to stop. I just started listening to this popular podcast recently, after meeting Marco at the Apple WWDC last June. Another one moving on.

Sitting at my kitchen table, looking at a print copy of the New York Times soon thereafter, I noticed something under the masthead: "No. 56,000". Fifty six thousand issues, starting with number 1 on September 18, 1851. Wow! When will it have its last print copy? Will it continue? I find I read an awful lot of the Times on my tablets or phones, with the paper mainly on Sundays. But I still think of it all as the New York Times. The paper version is starting to feel old, with the ink staining my fingers. As a printer's son, that's sad, too, but as a computer industry person, firmly in the mobile development world, it's also validation that I'm following the course where history is going.

Sunday, November 18, 2012 
Original VisiCalc running in a JavaScript IBM PC emulator [link]
Last Wednesday morning when I checked Twitter there was a tweet from Ray Ozzie (@rozzie) intended for me and Bob Frankston telling us to try out jsmachines.net with VisiCalc. I went to the URL and tried it: An IBM PC emulator written in JavaScript, with an easy way to load the copy of early IBM PC VisiCalc available on my web site.

To see what I saw, you need to let the emulator run -- boot up -- and press Return twice to enter the default date and time. Then select VisiCalc from the dropdown and load it as the "A:" drive. Then type "VC" as the command to execute it. Ray luckily remembered some of the program's commands so could operate it -- for the rest of you, check out the Reference Card on my site.

This is so cool. I really love JavaScript, having written a well-performing spreadsheet (SocialCalc) in it.  But here it's fast enough to emulate a CPU to run our original old machine code. According to the jsmachines.net site, it's throttled down if necessary to emulate 4.77MHz like the original IBM PC. This may become important for some apps (like VisiCalc's sound-making code) that depended upon the clock speed. However, if you set it to run full-speed, it can get to 10 times faster, depending upon your CPU and which browser you are using.

Something else cool this let me do: Run VisiCalc on an iPad, an iPhone, and an Android Nexus-7. When running on iOS, you can bring up the keyboard by tapping on the emulator screen. That didn't work on the Nexus-7 for me, but I was able to use a Bluetooth keyboard successfully. The iOS on-screen keyboard doesn't have the arrow keys, but the emulator has them as buttons.

Here's a screen shot:

VisiCalc running in Safari on an iPad mini
The emulator was written by Jeff Parsons (@JeffPar), currently a consultant. He's a 14-year veteran of Microsoft where he was a Software Design Engineer. He had previously worked on other emulator-like projects and drivers on other systems, so has a background quite appropriate for doing this.

Jeff tells me that he had seen some other JavaScript-based emulators and decided to see if he could make a really fast one. First he did a 6502 emulator of the OSI Challenger 1P. That went well so he decided to try the IBM PC. When he put it up on this new domain, mainly so he could test it remotely, it somehow become known publicly (a surprise to him -- he was out of town). It is still a work in progress. (For example, he says there are some bugs with the keyboard under iOS.) He may be releasing the emulator source under GPL at some point (his source files all have GPL notices he tells me) and is planning more documentation.

Thanks, Jeff, for helping bring the original VisiCalc to more platforms, keeping the historic software alive in the modern world.

Archive   Home   Newer   Older
© Copyright 1999-2014 by Daniel Bricklin
All Rights Reserved.

See disclaimer on home page.