Ever since I upgraded my iPhone to IOS4 (the latest version of the phone’s operating system), the Mail application has been flaky when it comes to syncing mail messages via IMAP. Duplicate messages, empty/blank messages, messages dated 12/31/1969, messages that are deleted and then re-appear, and so on.

At first I thought it might be my phone hardware, which had been cursed from the beginning (a story for another time), but after that phone died and Apple replaced it with a brand new one with fresh firmware and settings, and it STILL happened, I was convinced it’s the software on the phone.  Other people are having the same issue all over the place.  But it can be hard to make Apple believe this – said the Apple Genius Bar worker at the Apple Store in Chicago, “they’re probably all just using the phone wrong.”  Wha?

So I decided to look at the source code of the iPhone mail program to see if I could find the problem, but Apple doesn’t make that source code available.

So I decided to turn on the Mail program’s debugging options to have it output the results of the IMAP sync sessions, but it turns out the Mail program doesn’t have any debugging options.

So I decided to look at the log files on the phone’s operating system to see if they reported any software issues I could address, but actually I can’t look at those log files even if they exist because Apple doesn’t let you look behind the scenes.

So I decided to report the issue on the project mailing list that all of the iOS4 developers monitor, and it turns out that no such mailing list exists.

So I decided to go report the issue to Apple to see if they can help.

After submitting a case online, I was called back in 27 seconds (FAST!) and talked to Chris from Apple, who works in iPhone support.  He was very good at gauging my level of tech savvy and talking to me at an appropriate level, so he quickly confirmed that I’d already done all of the normal troubleshooting steps they recommend.  I asked him if this was a known bug, and he said not as far as he could tell.

Chris then connected me with Shawn who is a senior adviser on the iPhone tech support team.  Shawn told me that in fact I had identified a known bug with the iOS4 firmware.  VINDICATION!  He said that they’ve mostly been seeing it in gMail syncing issues, but that they’re aware of it as a wider IMAP issue.

I asked Shawn for the unique identifier number for the bug so that I could go follow the progress of a fix on their public bug tracking system, but it turns out that Apple doesn’t publish information about known bugs until they’re fixed.

I asked Shawn how I could learn more information about when this would be addressed or other recommended fixes in the meantime, and he said I would just have to wait and see.

(Chris and Shawn at Apple were excellent tech support reps within the confines of what they’re allowed to do and say.)

Bottom line: these are the horrors of using a closed-source, proprietary product for a key piece of tech functionality in my life.

What Apple can do better:

• Test its Mail software’s IMAP syncing functionality more thoroughly before releasing it to production.
• Publish information about known software/hardware issues BEFORE they’re resolved so that people having those issues can see what’s up and possibly even contribute to the resolution.
• Train its store Genius Bar workers so that they don’t dismiss a legitimate complaint about functionality problems with smiling condescension.
• I’m pushing it here, but hey, it’s my blog: release the source code for iOS and related applications.

If you’ve had this issue with your iPhone, or similar software/hardware troubleshooting experiences with other vendors, feel free to share in the comments.

Update 9/17/2010: After installing iOS 4.1 a week or so ago, all of the IMAP syncing issues seem to have gone away.  I’ve seen a few reports that problems remain for others, but it’s clear that this release included some fixes for the mail application that weren’t described in the release notes.

1) The Pal-Item reports on a meeting happening today about technology in schools:

The Richmond Community Schools Board of School Trustees has scheduled a special meeting for Tuesday to learn more about the school corporation’s technology needs…Technology coordinator Rob Tidrow has reported to the board that the school corporation is operating with outdated technology, and in some cases, technology that is obsolete…School officials have asked the board to consider paying for the upgrades with dollars available in a rainy day fund or the school corporation’s savings account.

I support our schools having current technology so that students can be engaged with and knowledgeable about how to get the most out of these tools. In the past that may have inevitably meant running the latest version of Microsoft Windows on the latest desktops from Dell (although there are plenty who would have said there were alternatives then too).

But in an age where many technology tools and services are online, and aren’t tied to a particular operating system or hardware vendor, it shouldn’t be a foregone conclusion that a school has to spend large amounts of money on proprietary software licenses and cutting edge hardware, when low-cost or free software and older but perfectly usable hardware can do the same job. In a time where school budgets are being cut, it’s worth looking at other options before reinvesting in more hardware and software that may also become obsolete as quickly as what they are replacing.

I pointed Mark to this issue, and in response he wrote up one particular technical approach that RCS could take. There are others, too.

2) Jason Truitt writes in his blog at the Pal-Item:

Expensive new technology requirements for 911 offices have brought six area counties together in a search for solutions. Fayette, Franklin, Randolph, Rush, Union and Wayne counties all use the same [911 emergency] system now, but it will be obsolete next year and no longer supported by Microsoft.

There may not be any off-the-shelf open source offerings that will meet this need, but for crying out loud, don’t just sign up for the next Microsoft contract only to find yourselves back in the same position a few years from now. While they’re teaming up, I hope they explore what other communities are doing to get more value out of their emergency systems, whether it’s creating systems themselves that use more open standards, finding new uses for old equipment, or even challenging the expensive requirements that might not be serving community interests as much as they are guaranteeing income for influential vendors.

These two particular cases may be decided as they always tend to be. I hope that in general, local decision-makers will realize that there are other ways to go that save money and make better use of existing resources.

I’m using a 1.5GHz Powerbook G4, which uses the PowerPC (PPC) architecture. I downloaded version 2.2 of the iPhoneOS SDK, build 9m2621-final.

After the regular installation of the iPhoneOS SDK (which is really just the regular Xcode install with some additional packages thrown in) was complete, the Xcode new project menu was notably missing the option to choose the iPhone platform. On the advice of this 3by9 article, I took these steps:

1. Mount the SDK installer disk image
2. Open the packages folder
3. One at a time, install all of the .pkg files that begin with “iPhone”, using the default install destination
4. When done, you’ll have a “Platforms” directory in the root level of your Mac’s hard drive. Move the contents of that directory into the /Developer/Platforms/ directory. If you’re like me, when you’re done, you’ll have three directories in /Developer/Platforms: “iPhoneOS.platform”, “iPhoneSimulator.platform” and “MacOSX.platform”.

Once that step is done, you’ll be able to launch Xcode and create new projects that use the iPhoneOS platform and application types. But, you still won’t be able to build or run those applications, because Apple has hardcoded the “i386″ (Intel) architecture as the only valid one to compile for. Thanks to a variety of comments on the above 3by9 post and comments 23-25 on this TUAW article, here are the remaining steps to address that:

• Close Xcode and edit this file (perhaps after making a backup): /Developer/Platforms/iPhoneSimulator.platform/Developer/Library/Xcode/Specifications/iPhone Simulator Architectures.xcspec
• Change line 12 to read Name = "Standard (iPhone Simulator: i386 ppc)";
• Change line 16 to read RealArchitectures = ( i386, ppc );
• Add this chunk before the last closing parentheses:

       // PowerPC
       {
            Type = Architecture;
            Identifier = ppc;
            Name = "PowerPC";
            Description = "32-bit PowerPC";
            PerArchBuildSettingName = "PowerPC";
            ByteOrder = big;
            ListInEnum = NO;
            SortNumber = 106;
       },

• Save the file and close it. Now relaunch Xcode.
• With Xcode open and your project loaded, right click on the project name under the “Groups & Files”, and select “Get Info”
• Under the “Build” tab and in the “Architectures” menu, change these options:
  Set Architectures to “Native Architecture of Build Machine”
  Set Base SDK to “Simulator – iPhoneOS 2.2″
  Leave Build Active Architecture Only checked
  Edit Valid Architectures to be not only include armv6, but also “ppc” and “i386″.
  (I think some of these may be redundant, but it’s the combination that’s working for me.)

When you next build your project, you shouldn’t get any errors about VALID_ARCHS.

I hope that’s helpful. Since this is only one particular combination of build hardware/architecture and SDK software version, I’m sure there are other combinations out there where the above may not work, your mileage may vary.

Now, back to working on building an actual application…

1) Apparently, all you have to do to throw off the facial recognition software that protects us from identity theft or worse, is smile:

The Indiana Bureau of Motor Vehicles is restricting glasses, hats, scarves — and even smiles — in driver’s license photographs. The new rules imposed last month were deemed necessary so that facial recognition software can spot fraudulent license applications, said BMV spokesman Dennis Rosebrough.

And then he had the gall to spin it as an improvement, since it would be horrible to admit that humans had done a better job:

The new technology represents an advancement of what the BMV already was doing, Rosebrough said. BMV employees always have looked at the old photo of a person to see if it looked like the person seeking a new license.

FAIL.

2) I was at a local video store yesterday, trying to rent a video using Anna Lisa’s account. I gave the cashier her phone number and name, and he said he’d have to call her to verify that it was okay for me to rent on her account. When she didn’t pick up, I offered to call her on my cell phone (in case she wasn’t picking up the call from an unknown number), and the cashier said, “okay, yeah, just ask her if it’s okay and then you can tell me what she said.”

FAIL.

Unzip the archive, then process the resulting files to obtain a numeric result. You’ll be taking the sum of lines from files matching a certain description, and multiplying those sums together to obtain a final result. Note that files have many different extensions, like ‘.pdf’ and ‘.js’, but all are plain text files containing a small number of lines of text.

Sum of line 4 for all files with path or name containing bar and ending in .xml
Sum of line 2 for all files with path or name containing bar and ending in .txt
Hint: If the requested line does not exist, do not increment the sum.

Multiply all the above sums together and enter the product below.
(Note: Answer must be an exact, decimal representation of the number.)

And my solution, starting from a Unix prompt in the directory where the files were unpacked to:

# find . -ipath "*bar*.xml" -print | xargs grep -h -n '.*' | egrep '^4:'| cut -d':' -f2
# find . -ipath "*bar*.txt" -print | xargs grep -h -n '.*' | egrep '^2:'| cut -d':' -f2

I then took the two lists of numbers, pasted them into a spreadsheet, and multiplied the two sums into the final answer. I started to look at Unix tools to sum a list of numbers passed as arguments, but unsure if Google was timing me, I opted for the spreadsheet instead to keep it fast.

I could have also used “cat -n” to generate the line-number-prefixed output, but for some reason grep was on the brain.

How would you do it?

I like this puzzle as a potential test for a network/Unix sysadmin, and plan to use it at some point (especially since we’re trying to hire a system administrator at Summersault). Maybe Google will release their puzzle generation code?

• Messages with a blank subject line
• Messages with a blank body, with only a mysterious attachment that’s supposed to explain everything when I open it
• Pretty much anything from Network Solutions / Verisign
• Anything that asks me to forward it to 11 of my friends for good luck
• Messages that start out by declaring that “this is not spam”
• Messages with more than three (3) consecutively-placed exclamation points in the subject line or greeting

What about you?

• Do you depend on Gmail or Google Calendar? Did you know they’re not ready for production use yet?
• The Rockridge Institute, a progressive think tank (THE progressive think tank for many) abruptly closes its doors because there wasn’t enough money coming in. But as a part of their exit, the description of “The Big Job” to be done is compelling, and so life goes on.
• I love a good idea I’ve never heard of before, and this place has lots of them
• Can you survive for 24 hours without your computer? I didn’t think so.
• One of the best YouTube videos I’ve seen in a long time: An Engineer’s Guide to Cats
• The Palladium-Item will be updating its website this week – and once again wiping out any archive of past reader discussions in their forums. I can’t decide if this is a blessing or a shame, and it’s probably both. I hope the new beginning represents a new mode of conversation, but I won’t hold my breath.

• Can I please have a paper record of my vote? The answer will likely be “no.” In Indiana, there is no way for you to have a written record of your vote. The state would like you to trust that it has properly received and counted your vote in its computer system. If there were a need for a recount, the state would use the computer record of your vote, not a paper trail that humans can look at. But hey, who ever heard of an election that was close, or one where a recount really mattered?
• Can you please show me a copy of the software source code that powers these voting machines? The answer is probably “no” again. The software that runs on the voting machines we use was developed by for-profit corporations, and the contract that your lawmakers typically make with them is that the source code won’t have to be available for the public to see. We just have to trust that they haven’t made any mistakes that might affect vote counts, or worse, any intentional changes that might alter the outcome of the election. But hey, who ever heard of someone trying to change the outcome of an election for their own benefit?
• Can you please provide me with a copy of the testing procedures to insure that my voting machine has not been altered or compromised? Many officials will provide you with a copy of the procedures if you ask in advance, and sometimes copies may be available onsite at the voting center. You may be charged a reasonable per-page copying fee. If you can’t get your verbal request fulfilled, submit your request in writing, which creates a legal duty for your public officials to produce the documents. Once you have them, you can decide if they’re sufficient for you to trust the democratic process to the integrity of the machines and the companies that make them. But hey, who ever heard of a government-managed process that wasn’t totally comprehensive and in the full interest of its citizens?

You can learn more about verified voting at VerifiedVoting.org and their page on Wayne County’s voting equipment. If you’d like to take further action on this issue, there’s a Citizens Tool Kit available that includes notes on how to raise awareness and make change in your community.

Here’s my configuration (perhaps mostly for geeks, but hopefully also for anyone who’s interested):

I started with an old 433MHz Dell desktop system with a 3.2GB hard drive that Mark had laying around, and beefed its memory up to a whopping 128MB. Other than that, it’s vanilla hardware, probably something you could even find in a dumpster behind your friendly neighborhood office complex. I plugged it into a battery-backup power supply, connected it to my local network, and stuck it in a closet where the noise wouldn’t bother anyone.

From there, I decided to use the FREE trixbox PBX all-in-one telephony system, because trixbox (which is based on Asterisk) has all the software you need to run anything from a simple home phone system all the way up to an enterprise-level call center. Installing trixbox on the hardware was amazingly easy – I popped the CD in, rebooted, and it did the rest. It took care of installing the operating system, the phone system software, and all of the other packages I could ever want. When it was done, it pointed me to a web interface that I could use to administer the system from there. I practically didn’t even need to have a keyboard attached, and I certainly didn’t have to mess with logging in on the command line and puttering around with configuration files. And, I get visual displays of call activity, e-mail and web access to voicemail, vital stats on the hardware, and lots more. Wow. Did I mention it’s free?

At Summersault, we’re using IP phones, which are basically phones that plug into the network instead of into a phone line. This is convenient because you don’t have to run copper wire everywhere (that’s a good thing to keep in mind if you’re trying to convert a high school into a national communications center for a Presidential Debate, eh?), and I could do that at home too and just take advantage of my local network setup. But, I decided that I wanted the flexibility of using regular old phones that I had laying around (especially the one with a wireless handset), so I bought a Sangoma A200 card with 2 FXS ports on it. That’s basically a fancy way of saying I bought a thingamajig that went in the computer that allowed me to plug a phone into it.

Then I had to figure out what to tell my system about how to route phone calls. In other words, when I pick up a phone and punch in some numbers, where should the call go? That’s part of the beauty of a VOIP system…I can set things up however I want. I could have an extension for the kitchen, and one for the basement, and maybe a shortcut that goes straight to a family member’s home number, and…the possibilities are endless.

But I kept it simple, don’t worry. Long distance calls are routed out through the CallWithUs.com service, which charges a flat fee of 1.38 cents per minute (that’s $0.0138, Verizon) to calls in the U.S., and I can refill my account there through quick and easy PayPal payments. No federal taxes, no start up fees, no monthly charges. Beautiful.

Local calls are a different story. I *could* route them through CallWithUs, but it doesn’t feel right to pay a per-minute fee for something that is traditionally “standard” in a normal phone setup. I could call Verizon and have a physical land-line put in (my Sangoma card has a port for that too), and just route all my local calls there, but then I’d have those pesky monthly fees whether I used it or not. The solution? For now, I’m telling my VOIP system to talk to the VOIP system at Summersault, and use one of the company’s available phone lines for outgoing local calls. Cool!

(Just think about all of the business phone lines that sit around unused during evenings and weekends, and how we could save lots of dollars by making better use of them, especially as a part of good local infrastructure planning! Mark and I have even joked about making “local phone calls” an employment perk, but we’re not sure that would be a real selling point, yet.)

Incoming calls are a slightly different matter, and I haven’t gotten that far yet. CallWithUs, Sipphone and various other providers like them offer dedicated local phone numbers for dialing in (sort of like you can get with Skype) for as low as $6.50/month, usually with a certain number of incoming minutes free. I’m not sure we need that yet, but however it might get setup, the features for handling incoming calls in trixbox are extensive. We could have it ring right through, or go into a menu system (“press 5 if you are calling to sell something…”), or go to voicemail, or forward to a cell phone, or….phew. One thing at a time.

So, that’s the setup for now. It’s definitely overkill for my simple home needs, but I enjoyed putting it together as a part of a proof-of-concept, and it’s neat (as a geek, anyway) to have that kind of control over every aspect of your phone setup.

There are lots of places to turn to if you want to try some of this yourself. I learned a lot of what I know about VOIP and Asterisk-based phone systems through my work in transitioning Summersault’s phone system to that setup. I can certainly recommend the books Switching to VOIP and Asterisk: The Future of Telephony (which, in addition to being available in sturdy book form, is also available free to download). They’re definitely geared toward folks who are ready to dive into the guts of telephony software, but they’re still accessible to those just wanting to learn more about phone infrastructure and what’s possible with VOIP.

And of course, through Summersault, I’m available to consult on these matters for home and business use, though I can’t promise that I’d feel comfortable charging for all of my time, since it’s just so much darn fun.

Update on 10/16/2007: I’ve now added an incoming phone number (DID) to the setup, also through CallWithUs. At $6.50/month with 2,000 free minutes included, it seemed like a great deal when the need arose. And it’s especially nice that they let you set your own outbound caller-ID, so all of the calls appear to come from the new number, making return calls even easier. Nice!

The main point is pretty simple: provide a real-time discussion room for Richmond/Wayne County citizens to talk about the issues of the day, chit-chat, and whatever else seems useful, any time, day or night. The way it’s set up now, a “robot” will periodically insert a headline, weather report, event, etc. from local sources into the room for those joined in to talk about. Users can carry on private chats with each other if they choose. Real names are encouraged, relative anonymity is certainly possible.

If it takes off, I plan to have moderated discussions on a particular topic, and/or moderated live interviews with local folks about particular projects/news item/etc., where those joining in could ask their questions directly. The same concept could be used for “liveblogging” an event like a City Council meeting, a Mayoral debate that sadly isn’t open to the public, an interesting speaker, etc. Eventually, the one chat room could expand into topic-specific rooms if desired.

While I recognize that online chat rooms don’t lend themselves to anything near the quality of exchanges as real-world encounters, my hope is that this format for online discussion might encourage behavior that’s more engaging, interesting and respectful than what happens in some of the traditional online forums available now. But who knows – it could go the other way too.

So, it’s an experiment.

Please feel free to join in (or not) and let me know what you think. The site is in “beta” but it should work fine; if you have any trouble joining the chat, just let me know.

(Yes, I’ve been here before, so perhaps I’m just repeating myself. But hey, if the Internets are good for anything, it’s inarticulately hammering away at the same vaguely relevant point in the name of personal amusement, yes?)