Enavigo

Why you should avoid Grohe bathroom fixture products

yuval — Fri, 20 Aug 2010 02:56:07 +0000

6 months ago, my wife and I were fortunate enough to be able to renovate our two bathrooms. The kids got a beautiful and simple bath arrangement while our master bathroom's shower was outfitted with several fixtures made by German bath fixture manufacturer Grohe. Grohe's brand name and positioning places it where you would find BMW in the car world totem pole. Expensive, great design and supposed bullet-proof reliability. Just to be sure, we purchased the fixtures from an authorized Grohe dealer in Waltham, MA, Peabody Plumbing Supply.

So it was truly disappointing that 10 days ago I stepped into our beautiful shower only to find out that the diverter valve, the part that switches between the overhead shower and the handheld unit, failed. Water was flowing from both showers, neither at a satisfactory flow. My wife called Grohe that morning to request a replacement part. The Grohe representative felt the problem sounded like a failed pin that is the main component in the diverter. He assured her the part will be sent immediately and we will have it within two days time. Just what I would expect from a reputable company.

10 days later and the part never arrived.

We reached out to Grohe 5 days ago to get an update on the part. No one was able to give us the information as the person my wife spoke with, Rick, was not around. My wife left him a message and also followed up with an email. No response and no part arrived to this day. This is now becoming very disappointing: my shower is disabled and the brand I put my faith in, appears to not care.

Where consumers finally enjoy some relief is in access to social media and I intend and already started spreading the word about my situation and how others should react. This is also why I am writing this post – you, strangers and friends, will hopefully now know that Grohe provides really crummy service experience (maybe they should read Putting the Service-Profit Chain to Work to understand why this is a big deal). Grohe appears to be on Twitter, but instead of listening for real, it appears to be a social media PR campaign (it's viral, according to their website) to share the joys of showering. Well, I cannot shower, can I?

My wife will now call them a fourth time tomorrow. We'll see how that goes.In the meantime I reached out to the campaign's Twitter account for help and to vent.

How to export contacts from a Nokia device to Gmail

yuval — Thu, 19 Aug 2010 01:46:35 +0000

I got my wife an iPhone this week. She used my old Nokia N95-8GB when we both decided we were fortunate enough that we could afford to get her an iPhone. The (virtual) keyboard and overall capabilities of the device just overwhelm compared to the Nokia.

Still, her contacts were on the Nokia and needed to be moved to the iPhone. Since her new iPhone uses micro-SIM cards, the Nokia's full size SIM could not (easily) be removed and moved to the iPhone. Software was our hope.

Google provides synchronization with mobile devices using the Microsoft Exchange ActiveSync protocol. To achieve that, the Nokia is expected to have Nokia's Mail for Exchange application installed. The Nokia had version 3.0.0.0 of Mail for Exchange (MfE) and sadly failed to sync no matter how many times I tried. Ironically, Nokia has a Sync application that works with their Ovi Contacts website. Sadly, Ovi Contacts does not appear to have an export capability for the contacts it holds.

So how do you do it?

My N95-8GB came with a software application on the CD called Nokia Nseries PC Suite. I believe it is similar to the Nokia PC Suite distributed today. I installed the application on my PC (no Mac version, sadly) and connected my phone. I started the application, which detected my device. I then clicked the tools button, and finally 'Nokia PC Sync'. PC Sync will then ask you if you wanted to synchronize your phone's calendar and contacts with Outlook, Outlook Express (which relies on Windows' Address Book installed on any Windows XP and newer machine) or variations of Lotus Notes. I chose Outlook Express as I do not have Outlook installed. The application then proceeded to ask me which address book to export to and then once confirmed, the application synced.

Once synced, I disconnected the phone from the computer and opened Address Book. The application immediately showed the new contacts imported from the phone. Under the File menu, Address Book has the ability to Export. Select 'Other Address Book' and then select 'Text File (Comma Separated Values)'. Pick a location on your computer and Address Book will create a .csv file there.
Update (8/19/2010): Apparently Windows Address Book on Windows XP, at least, does not export mobile numbers in its export function. To achieve this feat you need to install Windows Live Mail. Windows Live Mail required me to uninstall a previous version of the Windows Live Essentials suite I had already but after doing that, it installed fine. Windows Live Mail comes with what apparently became Windows 7's address book – Windows Live Contacts. Windows Live Contacts, in turn, reads Windows Address Book's files but is able to let you export the correct fields – most importantly, mobile phone number.

To do that, open Windows Live Mail and select 'Contacts' from the application bar on the bottom left hand side. The contacts window will open. Now, press ALT+M or click the menus button. Select 'Export' and 'Comma Separated Values (.CSV)' from the sub-menu. A window will pop up asking you for a file name and location for the exported contacts file. Select one and click 'Next >'. You will now have the option to select what fields or contact attributes you would like to export. Scroll down the list and make sure that the 'Mobile Phone' field is checked. Click 'Finish' and the application will create the CSV file you need in order to import your contacts to Gmail.

Finally, go to Gmail and log into your account. Click 'Contacts' from the menu on the left. Your contact list will appear. On the contact list menu, click 'More Actions' and then 'Import…'. A popup will ask you to select a .csv file. Select the one you created with the Address Book application and click the 'Import' button. Google will read the file and voila – your contacts will be part of your Gmail account. This account can in turn run synchronized with your iPhone.

Spring Bean Validation: Error code list for Custom Error Message

yuval — Fri, 13 Aug 2010 17:46:14 +0000

In my charge to regain my coding skills, I am using Spring MVC and trying to run it on Google's App Engine.

As with any web application, I have a form which I was looking to validate. Spring appears to be very high on its integration of Hibernate's implementation of a budding Java standard, JSR-303, In essence that means annotating a POJO bean file and then having your Spring MVC Controller (@Controller is the annotation) use another annotation (@Valid) to trigger validation. If errors are found, the form is returned with error messages. Pretty awesome. Yet JSR-303 has a rather big omission: if you have to fields in a form that must be identical – as in the case of password a password confirmation field, you need to write custom code. Luckily, that capability is covered by *the other* Spring-related validation framework – The Spring Modules project's Validation library.

Configuring Spring Modules' validation is covered insanely well by Willie Wheeler. Spring Modules Validation (SMV hereafter) has the ability to compare fields to each other as a validation condition and also apply a validation rule based on prior validation success (look here for table 17.11). Very very cool powerful stuff.

Where his article needs a bit of help is in how to set up custom error messages. See, SMV detects a validation error, it looks to create an error message that your Spring MVC form will display to the user. The default error messages are intentionally lacking, as you *are* expected to provide your own version for an error. And while the Wheeler article talks about how to set the errors up in a Properties file, he does not (nor is he expected – again thank you) cover all possible validation error codes that are necessary for you to define the error messages. This is why I am writing this post.

So, your custom error message properties file will consist of lines like the following:

SimpleClassName.propertyBeingValidated[validationRuleViolated]=your custom error message

So if you had a class called ApplicationUser with a field called email, your annotated class code will look this:

@Email private String email;

And your error message properties file will have a line like:

ApplicationUser.email[email]=Please enter a valid email address

You need to ensure that your class name (no package needed) is correctly capitalized. What's not clear – and drove me nuts, is what exactly are the error codes you put within the brackets – for example – [email].

Thanks to this incredibly useful post on the SpringFramework discussion boards, the error codes are based on default strings that are embedded in each one of the validation classes inside the org.springmodules.validation.bean.rule package.

To spare you digging through the source, here are the validation rules and the default error code strings associated with them:

Validation Rule	Error Code	Description
`@NotNull`	`[not.null]`	This rule validates that the validated value is not null.
`@NotBlank`	`[not.blank]`	This rule checks that a string value is not blank (that is, it holds some characters).
`@InTheFuture`	`[in.future]`	This validation rule checks that the validated date/calendar/instant occurs in the future (relative to the time of validation).
`@InThePast`	`[in.past]`	This validation rule checks that the validated date/calendar/instant occurred in the past (relative to the time of validation).Attributes are `value` that"Specifies the condition expression " and optionally, `expressionScope`. Defines the evaluation scope of the condition expression. global will define the validated object as the scope, while property will define the property value as the scope. Using the property scope enables the definition of conditions that apply directly on the property value (e.g. "length > 5"). The global scope enables the definition of conditions that apply on other properties of the validated object (e.g. "equals some_other_property")
`@Email`	`[email]`	This rule validates that a string value holds a valid email address.
`@Expression`	`[expression]`	This annotation represents an expression based validation rule. See Table 17.11
`@Length`	`[length]`	This rule checks that the range of a string's length is within specific bounds.See Table 17.13 for extra attributes
`@MaxLength`	`[max.length]`	This rule checks that the range of a string's length is less than or equals a specific upper bound. Must have a `max` attribute that "Specifies the maximum length the string value can have"
`@MinLength`	`[min.length]`	This rule checks that a string's length is greater than or equals a specific lower bound. Must have a `value` attribute that "Specifies the minimum length the string value can have"
`@NotEmpty`	`[not.empty]`	This rule checks that a collection/array is not empty (that is, it holds at least one element).
`@size`	`[size]`	This rule checks that the size of a collection/array is within specific bounds. See Table 17.16. Essentially, `min` and `max` attributes
`@MinSize`	`[min.size]`	This rule checks that the size of a collection/array is greater than or equals a specific lower bound. Must have a `value` attribute that "Specifies the minimum size of the collection or array ".
`@MaxSize`	`[max.size]`	This rule checks that a `java.lang.Comparable` value is within a specific range. Must have a `value` attribute that "Specifies the maximum size of the collection or array".
`@Range`	`[range]`	This rule checks that the size of a collection/array is less than or equals a specific bound. Must define either a `min` or a `max` attributes or both. Specify the lower and upper bounds of the range, respectively.
`@Min`	`[range]`	This rule checks that a java.lang.Comparable value is greater than or equals a specific lower bound. Must define a `value` attribute that "Specifies the lower bound of the range"
`@Max`	`[range]`	This rule checks that a java.lang.Comparable value is less than or equals a specific upper bound. Must define a `value` attribute that "Specifies the upper bound of the range"
`@RegExp`	`[regexp]`	This rule checks that a string value matches a specific regular expression. Must define a `value` attribute that "Specifies the regular expression"
`@Expressions`	`[expression]`	This annotation enables defining multiple `@Expression` annotations on a property.
`@ConditionRef`	`[spring.condition]`	This rule uses a condition that is defined in the application context to perform the validation check. Must define a `value` attribute that "Specifies the name of the condition bean as defined in the application context."

Hope this is of help to others.

Charter school vs. Old school

yuval — Mon, 31 May 2010 18:52:41 +0000

Great story from the Wall Street Journal about the diverging paths of two Oklahoma City students. The article aldiscusses the effect their choice of high school had on their future. A great read.

http://online.wsj.com/article/SB10001424052748704717004575268752238805736.html

The Motorola Dilemma; do agencies have a lesson there?

yuval — Mon, 31 May 2010 01:52:24 +0000

A Wall Street Journal story about Motorola's relationship with Verizon and Google yesterday made me think of the tough situation mobile phone makers are in. The company is betting on partnerships with Verizon and Google for its future. Verizon as a carrier is important, but Verizon will go wherever the next iPhone wannabe is, not caring who makes it. Google, on the other hand, is where the risk lies.

Using a common operating system is nothing new. Nokia's Symbian was used by many manufacturers successfully for almost a decade. But this common foundation, while breeding a 3rd party app developer ecosystem, drastically increases the demand to produce great hardware. If your device cannot cut it, consumers will reject it. The alternative is to ditch the common operating system and create a combined, premium software/hardware combination. Apple succeeded, Palm did not, Samsung dreams it can. In Motorola's case, its own old basic operating system survives on low-end phones, while it realizes savings from avoiding its own Linux-based operating system's development costs.

In a way Motorola has something a reason for optimism. It DID produce excellent designs in the StarTac and Razr series. But now, if the hardware is not compelling enough, it will be toast. Worse, it is relegated the ranks of generic hardware manufacturers like HTC (well, they wish they were HTC at this point), LG, Samsung and Pantech. So the key here is design. And design translates to a unique proprietary flavor. The gist of 'why would I buy this phone as opposed to the others'.

Is this a unique business situation in a specific marketplace? Not really. Retailers compete to sell the same product on shelves they think are better organized, in stores they hope are cleaner and providing service that excels beyond that offered by their competitors. Technology offers a level playing filed in creating something that a brand can say is unequivocally superior. Snappier operating system experience is one; better video playback, a great keyboard or camera, remarkable call quality and stability are others. Again – proprietary flavor, a unique business proposition, the differentiator.

Working in an agency we operate in a miasma where beloved clients who love you have 6-18 months on average in a job. The next guy will bring his own agency and you're gone. Interchangeability is the name of the game and being on par with your peers is a good and common place to be. That is the core of why the problem is similar: if your agency lacks differentiation, you will be considered, invited, known, but nothing will pop to people's minds when they think of you. If you have a good idea that services the client well, great; you might get the assignment. What is missing is that differentiator, and I look right back at that concept of 'proprietary'. Proprietary is the hook. Proprietary is what you can do, only you have access too, and is a factual advantage to you. And when the client likes that proprietary capability/service/offering – they have no choice but stick with you.

For an agency that might mean having creative that knocks the ball out of the ballpark every single time. But that's difficult to achieve and can be draining. I am a technologist. I care about building stuff. And I think an agency needs to invest (!) in building technology that is the differentiator, maybe even the foundation, for its relationship with clients. That's really challenging, and may be expensive, but can simplify customer relationship lifecycle – acquisition, retention, expansion – from the current rat race. It is a very steep slope to crawl upon. You need to discover a need, do it once, and bet on repeating it. Worse yet, the agency suddenly needs to support technology, not only customer needs. Having a product is an entirely complex ball of wax all by itself.

Do agencies have a choice to avoid it? I am not sure. You can still partner with vendors – like Motorola does with Google – but your competitors can do so too and come up with a better offering that you will. Brands are not stupid either. They are happy to go straight to startups to get their own edge and to avoid paying agency overhead.

Therefore, agency folk, the choice is yours: become more of a startup, get your proprietary flavor, or remain adrift in a very flashy variety of a sea of gray.

Nvidia, Intel and Ubuntu: Fail? [SOLVED (again)]

yuval — Sat, 29 May 2010 17:24:04 +0000

Let me start by saying that I am a fan of Ubuntu. I think the project is a huge success in making Linux digestable for the masses and taking away the oodles of nonsense that sucked time away just getting a machine up and running. Issues that I had with a built-in graphics chip lead me to buy a new video card. And that's when the headaches started. Luckily, I think I finally solved it…

I recently received an old desktop machine from work. It is an HP Compaq d220mt. A solidly built box with a built-in graphics card from the Intel i845 chipset. I wanted the machine primarily to be used by my 4 year-old son for games online. Again, Ubuntu was simple enough for a 4 year-old to use (more of a success for Xfce!). Still, whenever he would play one of the Flash-based games on PBSkids.org, the machine would suddenly hang. Not a great experience. Upgrading Ubuntu did nothing to improve the situation.

Looking to avoid more time waste I decided to try and get an alternative video card for the machine. NVidia supposedly has decent Linux support so I bought the PNY GeFirce 5200 graphics card that uses the PCI slots available on the machine. Not AGP, not PCIe; old-school PCI. Cool card with dual monitor capability and all and a whopping (for me at least) 256MB. $30 before rebate.

To be sure the machine is ready, I did a clean install of Ubuntu 10.04 using the built-in Intel graphics chip. I then installed the new NVidia card. Restarting the machine after switching to the PCI graphics card in BIOS lead to a crash, and the machine failed to boot. I then switched back to the built-in Intel graphics and the machine booted. I installed the proprietary drivers from NVidia using Ubuntu's special tool. Restarted the machine, switching BIOS back to PCI graphics. Again, a crash on startup. Switched back the BIOS to the built-in chip and Ubuntu intelligently noticed the crash and told me that the X Windows configuration was off. It let me review it, modify it (I added the specific PCI address for the NVidia-based card after checking the boot log), but there was no clear way to save the configuration file. Restarting the machine with the NVidia card (PCI graphics) back on (again switching the BIOS), lead to a crash.

I tried another tack: saw some posts talking about a startup setting through the GRUB boot loader of 'nomodeset' that will force Ubuntu to startup with failsafe graphics. That did not work.

Tried reinstalling Ubuntu 10.04, with or without the 'nomodeset' setting. Crashes early, leaving the Ubuntu startup screen dots flashing. Tried installing Ubuntu 8.04 LTS. Here Ubuntu crashed again before getting me to any of the option screens. Live CD fails as well on either version.

Disappointed, I tried Fedora 13; crashed virtually on startup. Centos 5.5 – crashed virtually on startup. Tried PC-BSD and guess what – it installed SEAMLESSLY. "Excuse me, WHO?" you would ask. PC-BSD, that's who. It detected the NVidia card, installed itself and restarted, asking me to adjust its display settings.

I assume that Ubuntu leaves the proprietary NVidia drivers out, but failing to even get to the initial install screens is truly disheartening. Ubuntu spoiled us with seamless installs that 'just get it done' (sorta like what PC-BSD did). As a result, I expect better from Ubuntu. It could very well be that my configuration is cursed, but PC-BSD got so far as to offer me the option of enabling dual-monitor setup, available on my graphics card.

I am still trying to reinstall Ubuntu, mostly because I really dislike the KDE desktop that is the only option with PC-BSD. If I find an answer, I will post it here. Stay tuned.

SOLUTION

I think I have it working fine now with Ubuntu 8.04 LTS. Amazing!

Refusing to let go of the issue, I decided to try another tack from a post I read somewhere and sadly cannot remember. The idea was to go an adjust the xorg.conf configuration file manually. Anyway, this is what I did:

Reinstalled Ubuntu 8.04 from scratch using the built-in Intel graphics card (turned it on in the BIOS).
Once installed, I booted into it and logged in.
Installed the NVidia graphics drivers: From the top bar – System -> Administration -> Hardware Drivers. Accepted the warnings about proprietary drivers and installed the new driver. Did not reboot yet.
Ran the package manager to install all security and software updates.
Opened a shell window and made a backup copy of the X Windows configuration file, /etc/X11/xorg.conf – for example naming it /etc/X11/xorg.conf.bak-yz. (Please look at the "Avoiding the mess" section below on how to avoid the mess that happened to me)
Rebooted.
Changed the BIOS to use the PCI graphics card instead of the built-in Intel chip.

Now this is where I had a minor pickle.

When restarting, the system managed to boot but instead of starting, the system started a failsafe graphics mode alerting me that the X Windows setup was weird. The configuration still showed (despite the BIOS disabling it) the built-in Intel chip. There was no way to remove it. Ubuntu also thought I was using two monitors with two cards. I tried changing the configuration using the tool but when I told Ubuntu to continue to boot, it failed and got into console mode again. This was actually a good thing. I was able to log into the system from the command line (and you can also use Alt+F1/Alt+F2/Alt+F3 to switch to other consoles (Unix, FTW).

On the command line, I looked at the /etc/X11/xorg.conf that was generated for me (vi /etc/X11/xorg.conf will do it). It indeed reflected the two graphics card – the built-in Intel one (that was now disabled) and the NVidia one. It also assumed two monitors, and well – it was incorrect. Running the dpkg tool, as mentioned in the body of the xorg.conf file to regenrate a new configuration only. resulted in a very bland non-configured xorg.conf file. This is where the backup of xorg.conf file came into play. So how do we finally fix it?

Part of the issue stems from the fact that Linux can still see the original Intel graphics chip on the motherboard when it prods the PCI bus. It sees the NVidia card too but it gets confused. You need to manually tell xorg.conf which PCI device to use for the graphics card you want. To do so, we need to add a BusID parameter to the Device section of the xorg.conf file – telling it which device on the PCI bus to use for graphics. To find the PCI bus ID I ran the following command:

sudo X :1 -scanpci

On my machine I found the NVidia card on the last line:

(3:13:0) unknown card (0x196e/0x01ad) using a nVidia Corporation NV34 [GeForce FX 5200]

The crucial number here is the first one: 3:13:0 is the PCI bus ID for the nVidia card. Note that the id MAY look different on your machine.

The next step is to use backup xorg.conf file, opening it using the command:

sudo vi /etc/X11/xorg.conf.bak-yz

This file had the correct configuration for a single video card, single screen X Windows setup with the NVidia card. I now scrolled down to the device section which looked like this:

Section "Device" Identifier "Configured Video Device" Driver "nvidia" Option "NoLogo" "True" EndSection

We now need to add the BusID parameter to this section. In my case, it ended up looking like:

Section "Device" Identifier "Configured Video Device" Driver "nvidia" Option "NoLogo" "True" BusID "PCI:3:13:0" EndSection

I now overwrote the xorg.conf file by clicking ESC and entering the vi command

:w! /etc/X11/xorg.conf

Update: There is one more step necessary here. Apparently Ubuntu resurrects the Intel driver. On Ubuntu 8.04 you can add the Intel driver to the machine's module blacklist. To do so, open the file /etc/modprobe.d/blacklist using

sudo vi /etc/modprobe.d/blacklist

and add the line

blacklist intel_agp

to the end of the file. Then save it and exit vi.

On Ubuntu 10.04 things are a bit more complicated.

Finally, I rebooted the machine using

sudo reboot

Amazingly enough, the machine rebooted successfully using the nVidia card!

Quite the hack, isn't it. Hope it is of help to anyone. It was not fun.

Avoiding the Mess

What I would recommend to you to do once Ubuntu generates the new xorg.conf file after installing the proprietary drivers is to open xorg.conf and to add the BusID section to its device section. Again, to do so:

Find the nVidia video card's PCI bus ID
sudo X :1 -scanpci

On my machine I found the NVidia card on the last line:

(3:13:0) unknown card (0x196e/0x01ad) using a nVidia Corporation NV34 [GeForce FX 5200]

The crucial number here is the first one: 3:13:0 is the PCI bus ID for the nVidia card. Note that the id MAY look different on your machine.
Open xorg.conf
sudo vi /etc/X11/xorg.conf
Add the BusID line to the device section, so it looks like

Section "Device" Identifier "Configured Video Device" Driver "nvidia" Option "NoLogo" "True" BusID "PCI:3:13:0" EndSection (REMEMBER TO USE THE PCI ID FROM STEP 1)
Save the file.
Blacklist the Intel driver: On Ubuntu 8.04 you can add the Intel driver to the machine's module blacklist. To do so, open the file /etc/modprobe.d/blacklist using
sudo vi /etc/modprobe.d/blacklist

and add the line

blacklist intel_agp

Now save and exit vi.
Restart.

Hopefully – not promising anything – this will help you avoid the problem I encountered.

Crossing fingers.

Geo-tagging photos on the iPhone: feature added, killed.

yuval — Mon, 17 May 2010 09:36:42 +0000

I love taking photos with my phone. Phone cameras turn you into a mobile content generation unit. Snap photo, upload, the world knows. Never mind the fact that the world knows about your cat being cute or your son's nose being very congested. Phone cameras become double powerful when coupled with the phone's GPS. The GPS stamps each photo's EXIF data, the same metadata that records when you took the picture, with where you did it. So in essence, you need two elements: upload and location-recording capabilities. My Nokia N95 had both capabilities. Great camera, GPS, uploads to Flickr and Nokia's Ovi. Then Nokia stopped developing apps for the phone, which is very disappointing. Its general slowness and the arrival of the iPhone 3Gs convinced me to move on. The iPhone compensates for middling camera hardware with speed and processing power. And it had a stellar app called PixelPipe. PixelPipe batch uploaded my photos and videos to any site I wanted. And little did I appreciate it at the time, but it also retained and uploaded the geotagging data from photos. Pixelpipe was recently removed from the iPhone's app store. Apparently it accessed photos using the wrong element (in Apple's eyes) of the iPhone SDK. That made the app better for batch uploads, but played against 'the rules'. So I was left looking for alternatives. Without mentioning all of them, Flickr's app is the most disappointing to me. Flickr's app's beautiful design, sensible usability, stable, but with one (probably imposed) 'feature' bug: it tags whatever photos it uploads with the location of *the upload*. In other words, if you took the photo in Japan and uploaded it in Omaha as- Mt. Fuji will be geotagged as being in Nebraska. Now why would Flickr, a great Website I love and pay for, do such a boneheaded thing? Because apparently that's the best the iPhone allows. It seems like the iPhone SDK forces developers to access photos using specific APIs that remove the location data from images. If you actually take a picture and then upload it immediately using the Flickr app allows the app to append the currrent location back to the photo. Absurd, no? So what's left? iPhoto and other desktop applications that get the actual phtoto files from the iPhone can still get the location data from the actual files. But on the iPhone, Apple made the wrong decision to remove that information. Privacy may be the concern. I am, for one, disappointed and a bit angry. I doubt Android imposes such a limitation on apps. But to follow Steve Jobs' logic from an email, I'd better create than criticize.

UPDATE:

Manage to restore PixelPipe from iTunes. It is the good version that was removed by Apple. Amazingly enough, having tweeted about it, PixelPipe asked me to send them their own iPhone app file. Probably for use with jailbroken iPhones. A bit surprising that they, of all people, will not have an old version of their own app. Still, glad I solved it by going around Apple's own restrictions. Geotagging is back for me.

Fill Haiti Love

yuval — Sat, 30 Jan 2010 04:02:41 +0000

iPad: Did the laptop just die [A Facebook discussion]

yuval — Fri, 29 Jan 2010 04:09:25 +0000

I want to believe that I know (ok, encountered) a variety of smart and interesting individuals in my life. Apple, as an amazing brand, warrants a lot of emotion from its followers. The release of the Apple iPad caused so much excitement and discussion that it hobbled Facebook and Twitter. So when I set my Facebook status message to 'Did the laptop just die?' in reaction to the iPad's launch, a tsunami of insightful comments ensued. I want to share it with the world as a snapshot to be revisited after common mortals get their paws on the device and we can judge it ourselves. (Persons names shown in acronyms)

YZ(Me): Did the laptop just die?

BS:
Not unless you like a device that can only do one thing at a time, doesn't run desktop application, doesn't have a replaceable batter and can't run flash
Yesterday at 2:32pm ·

TR:
Maybe they should have come up with a name that doesn't evoke images of a feminine hygiene product.
Yesterday at 2:33pm ·

YZ:
You want to tell me you do not use a sanitary slate?
Yesterday at 2:33pm ·

RWK:
Not since I got an iUD! Badoomchoo
Yesterday at 2:42pm ·

MB:
That was the sound of the now clearly overpriced Kindle dying.
Yesterday at 3:10pm ·

CA:
Don't forget that it can only run approved applications, and view approved content. Who cares how "shiny" it is, or how it can (or can't) multitask, if you can't do what you want with it? I wouldn't pay $5 for it.
Yesterday at 3:25pm ·

YZ:
I think Apple calculated the loss of the CA: population. They will go ahead and sell it anyway.
Yesterday at 3:26pm ·

TR:
The Kindle is half the price of the iPad and maybe some of us just want to carry a thousand books around. Apples and oranges comparing the two, same form different function.
Yesterday at 3:29pm ·

MB:
Half the price: single function, monochrome, passive display, no multi-touch. The market will determine whether the Kindle has a future at half the price, but to me it looks like a lot less than half the value.
Yesterday at 3:36pm ·

YZ:
I am starting to like this new 'agreeing with MB:' thing.
Yesterday at 3:37pm ·

TK:
iPad will sell like hot cakes, however I am really disappointed you cannot draw onto it with a stylus / pen. That would have made it extraordinary for me, and probably changed the way designers work. Also, no multitasking is crap. Someone will jailbreak it though, and make it a whole load better.
Yesterday at 3:41pm ·

CA:
Why would that be revolutionary for designers? Wacom (and others) have made devices like that for almost 2 decades.
Yesterday at 3:44pm ·

MB:
Two words, CA: direct manipulation. Wacom tablets are almost as disconnected as mice. The result of you intent does not show up under your stylus but on a separate screen. Had Apple allowed for use of a stylus, we would finally have a digital canvas.
Yesterday at 3:50pm ·

CA:
I understand the difference between a tablet and a touch screen (and can see the advantages) – and Wacom (amongst others) makes both. For example, checkout the Wacom Cintiq series, HP's TouchSmart, 3M's single and multitouch displays, MagicTouch's product line, etc. Why would an Apple device be a "digital canvas" when other devices (even those that work with MacOS) would not be?
Yesterday at 3:58pm ·

TR:
Maybe I don't want something that's multi-function, technicolor, and interactive, because I've already got a MacBook. I don't buy things because they're cool. I buy them because they work and do exactly what I want them to do – such as carry around a thousand books. If I'm carrying a MacBook, a Kindle and an iPhone, why would I add an iPad? "Mini" laptops have generally failed so far. Could the market already been too crowded?
Yesterday at 4:10pm ·

TK:
Cintiqs have poor resolution, and are expensive. If iPad has same res as the iphone, it would be very sensitive and really nice to draw on.
Yesterday at 4:11pm ·

MB:
Interesting, though design certainly matters, I don't think I factored "cool" into the price calculation.

You wouldn't add an iPad to your trio: you'd replace your Kindle (period) and MacBook (when you are not at your desk).

Lastly, I think the value of large screen, affordable multitouch UI has yet to be explored. I tried the Kindle but … See Morecould not get past the controls. The iPad solves that issue. Not sure whether I'll be version 1 (because I also already have a MacBook and an iPhone), but I can see the attraction.
Yesterday at 4:54pm ·

NC:
sorry MB, you'll be disappointed…. What everyone seems to forget is half of what makes the Kindle a solid product is the e-ink. Without it, iPad users are just BEGGING for migraines… You'll be reconsidering the Kindle (or Nook) in no time… iPad === Newton === FAIL (Though we all agree it will sell a ton; then again, didn't the Phantom Menace sell a ton of tickets too? )
Yesterday at 5:12pm ·

MB:
Maybe, NC. I stare at my computer screen for hours every day with no migraine. How is this different?
Yesterday at 5:30pm ·

MB:
Also, as someone pretty attuned to user experience, I really felt that the Kindle must be succeeding *despite* the ui. I used it for a month and returned it.
Yesterday at 5:34pm ·

NC:
Come on MB, you've seen heatmaps, you know that reading eye movements differ from typical computer use… Eyestrain ensues.
Yesterday at 5:43pm ·

NC:
btw, I will concede that the UI on the iPad will be better than a Kindle. I'm saying that the core reading will be far worse…
Yesterday at 5:44pm ·

MB:
I'm sure individual mileage will vary, eye strain wise. Clearly, when reading is the primary purpose readability is paramount; I just found the UX (UI + controls) of the kindle to be a dealbreaker.
Yesterday at 5:50pm ·

TK:
Apple is well on the way to dominating the mobile devices market, not only through the device itself but the UI. All these systems share similar interfaces- total compatibility. No need to learn anything new, it's second nature. This has been Apple's strength always.
Yesterday at 6:04pm ·

TR:
I agree with everything NC says. Also, am I the only person to have owned three touch-screen devices and hated them all? I'd rate the Kindle the easiest device I've ever learned to use – tremendously easier than the iPhone. Maybe I'm just too old-school for the paradigm. Also when I'm on 10-14 hour flights – as I oftem am – I could never read books on a computer-style screen.
Yesterday at 6:14pm ·

MB:
TR, I think the UI paradigm of the Kindle and the iPhone/iPod touch/iPad are pretty diametrically opposed.

Also, I think nailed the ultimate difference between the iPad and the Kindle: the Kindle is (overwhelmingly) a single purpose device: reading. iBook on the iPad is just one other application, not the primary purpose of the device.

However, I do think the iPad will dominate the "ereader" market in the education space in a way the Kindle and Kindle DX hasn't.
Yesterday at 6:21pm ·

TR:
Apple isn't on the way to dominating the mobile device market until they make a phone that's accessible to the 95% of the world that can't afford to buy an iPhone. They're on their way to dominating the smartphone market, but Nokia is still making a million cheap phones *per day*. Do you really think Apple is going to start supplying cheap mobile phones to Africa and China? No moreso that you'll see Ferraris on the roads of the Sudan.
Yesterday at 8:28pm ·

TR:
For anything to dominate the eReader market in the education space (which is a tiny, tiny market), textbooks would actually have to be available in eBook format. Very few are. It's not a device problem, it's (if I may be so bold to say so) a format and content management problem. The worst thing about owning an ereader is that so many books aren't available in the format.

The best UX rareley makes for the most successful product. There are so many other factors.
Yesterday at 11:02pm ·

MB:
An excellent user experience ALONE rarely makes for the most successful products, but an awful user experience can ruin a product's chances all on its own.
Yesterday at 11:06pm ·

TR:
Most successful products have mediocre/passable UX, and succeed based on other factors. Top example would be the Blackberry.

What do you think?

Favorite smoothie

yuval — Sat, 16 Jan 2010 17:09:56 +0000

I am not a big fan of fruit. I was just raised that way with parents who never insisted on me eating healthy. I still understand the importance of fruit and was always a big fan of Jamba Juice. After moving to Boston I left Jamba Juice behind and started fending for myself, experimenting with a variety of fruit combinations.

The first thing you need is a good blender. Blenders are quite cheap and $35 buys you a decent one. I love Osterizers.

I also use soy milk. Better for you, no cholesterol and lower risk of heartburn. Still plain milk is just fine. On the same vein, I used to add ice cream to my smoothies for extra flavor and chill. Then I moved to just using ice cubes and surprisingly there was little difference while lowering calorie count and fat. Just make sure to crush the ice.

So now, finally to my favorite smoothies combo:

- 3 ice cubes
- 1 banana
- 4-6 strawberries
- handful of cantaloupe
- 2-3 cups soy milk

Blend. Drink. Enjoy!