Thursday, April 19, 2012

So why webOS failed?

I have just had a fresh look into webOS, and why it failed.

It is too open, too neutral.

Why would providers "forget" about their preferenced partners?

In webOS, twitter is just a "Service", and to twit is to perform a "webOS Action".

There's no point to put your brand out, it will still become a subdued "part of" webOS.

Now if you're Google, Twitter or Facebook major internet (web) brand, would you want to step forward, and put your own "app" out, or would you be resenting to an almost empty corner of hp Catalog, the "Services & Actions" corner?

Users - they too, are so accustomed to what they are "fed", according to the iOS 'successful' agenda?

Tuesday, March 6, 2012

NOKIA will stop "sharing"


Just in the e-mails -
on May 30th, NOKIA will shut it's Ovi Share service and will delete any data you may have accumulated within.

Anybody surprised?

Despite being conveniently integrated into Symbian, the service did not get traction, mostly because it was available exclusively on Symbian mobile paltform.
Or your choice of a PC...There was nothing for iOS or Android.

Since NOKIA is walking hand-in-hand with Microsoft, they will probably be incorporating Microsoft's "sharing" services into future products.

As for Symbian, I guess the best choice is get a facebook plug-in from furtiv. Or use flickr.

Does anyone care? Chime in - what shall you use next?

Friday, March 2, 2012

"Windows" 8? Wait a minute


Microsoft Windows 8 is a very brave endeavor on Microsoft's part.
It essentially aims to replace, revolutionize the user interface that people use to "compute".

Wait a minute... "Compute"? Well, it seems that this term is obsolete, too.

Most of the times people don't crunch numbers, they are mostly after consuming numbers.
Modern PCs performance depends heavily not only on the pure computational power, but on storage transfer speed, network speed and memory size (even if that's video memory).

Did you notice Microsoft's stubbornness applying the "Windows" moniker to all their Operating Systems?
Is not it time for a product name change, as opposed to just the number?

Effectively, on a mobile device, we would not have the Desktop metaphor as invented by bright minds of Xerox PARC, and later implemeted by Apple Computer Inc.
Hence, we won't have overlapping "windows" (think "paper sheets"), representing different applications.
The "fourth screen" is too small, to have many different windows. We would have one window (possibly with many panes, as in highly acclaimed alas mismanaged webOS). Or multiple "tiles", replacing iOS's boredom of icon grids.

So they are not windows anymore, but "tiles". So why not "Tiles 1.0"?
Certainly, Microsoft thinks about itself in a very serious tone. They know they have to be brave, but consistent naming gives users false impressions, that the next OS is the same old, but better.

Ask many, and they will opinion - it's the new world, but not necessarily "better" in all senses. It's "different", and time will tell how quickly it will improve and be accepted.

Microsoft has learned to "think different" after all. Kudos.

Now, to the essense of what's going on.

Intrinsically, laziness wins, even when there are other grave considerations against it.
Did you notice, there's no ideology anymore, that would foster utmost strive for self-improvement and creativity in masses? The world is driven around money, not high ideals to achieve. Money is a carrot, it is also a stick.
With Information Superhighway (Bill Gates, 1995) being built on scale, and penetrating every house, one important thing has happened.

Digital Communism.
It has arrived. There is a mechanism, a technology invented and machinery built that makes [some parts] of former social regime obsolete. DMCA, ACTA, PIPA, SOPA, ring a bell? They are trying to reing it in, still wrap it under Capitalism, but "times they are a'changing".

When a major technological breakthrough happens, the social regime has to change.
Remember feudalism? Once a certaing advances have been made in producing mechanical machinery and steam engines, electricity were invented, these allowed material goods to be produced in quantities (sic!) that led to major changes in quality of life. This became Capitalism in most parts of the world, some have tried to build Communism, and very few have built a Socialistic-like societies (thank you, Sweden). There's no more feudalism - at least globally.

With Internet giving people "near-free" access to information, another social change is happening.
We have recently seen a wave of resistance from "undemocratically chosen" lobbyists pushing laws while the general public was asleep.
One big thing not solved yet - how to make new things "near-free" for everybody, in a decentralized manner.
But hold on, nanotech is coming.
Once the fundamental problem of easily "copying" material things verbatim is solved, "goodbye, Capitalism".

When it comes to gadgets and widgets we use, I observe a worrying trend.
A Personal Computer used to be thought of as a "bicycle for a mind", an open-ended thing that you would have full liberty to program, assign to any task you'd wish.

Nowadays, PCs are being replaced by a "shopping cart for a mind", a TV with many channels.

Most people who use computers now (as opposed to 70s and 80s) have no capacity to program them, they are formed by the Capitalistic regime to absorb what's being fed via "channels". Those who are not lazy, are building their own "channels", and take part in existing social mode, but on a different side.
Ironically, it's a bicycle which is preset to so many roads; you could always buy more "routes" for it, but all routes have to be approved, no riding around Whtie House anymore.

The conceptual idea of SmallTalk OOP language - to make PC programming so easy as to let anyone build programs for it - has been betrayed, privatized and forgotten.

The big mainstream capitalistic companies will have very strong control over all developers, very soon.
They will enforce digital signatures, Trusted Computing, and will revoke access of developers, following only their own internal company policies, not the government laws that cannot interfere with go-to-market strategies of private companies.

When Microsoft's transition to Tiles complete, we will only install what Microsoft likes, or allows us to install. Everybody in Metro development will have to be on good terms with Microsoft.
Microsoft may take a policy of accepting only "prudent" developers.
Doesn't it remind of something?

Customization options will be limited.
Did you notice it's not possible to use own color schemes in UI of major OSes?
Back in XP/2000/98/95 era you could paint windows your own colors.
Soon, even Wallpaper will go way of Dodo. Your only consolation would be to go buy more new tiles.
Reminiscent of Zamyatin's "We".

But most importantly, TV is back with vengeance. Just pick your channel.
Oh and what a TV is it.
It's a TV that watches you, knows your moves, your Contacts and Calendar.
The vendors only have to offer free "convenience" of backup for all of these.
Maybe it won't be offered, but built-in, mandated.

A lifetime account, anyone? The government would create it for you as you're born.
When you grow up to buy your device, just apply your RFID tag in your forehead and logon. A new ritual.
You will be able to apply to transfer it from MS to Apple or Google, but this will only increase number of parties who know a lot about you.
The irony is that Macintosh was started with the ad denouncing exactly the system they have been so instrumental in building.

I know you must be thinking I'm over-dramatizing it now.
Yes, I am. Some of it may not happen. Only good things will happen if every user's aware and informed, and uses own judgement.

Just remember, Gods live in the Clouds :)

Friday, January 13, 2012

To XenApp or not to XenApp?

There are few tips about Citrix XenApp (ex. Presentation Server, ex. WinFrame) that I would like to share.
Many people, even being involved in technical architecture decisions, are not completely aware of possibilities, dangers and ramifications of this product. They fall to "typical config" patterns that are outdated and do not deliver the best of XenApp.

1. Only publish publishable.

Sometimes XenApp is used as a Remote Access Solution. In this mode, they try to use XenApp as a sort of "gateway" into other internal systems. Well here's the gotcha - if you cannot install and publish a client on XenApp's farm, do not use XenApp. I have seen deplyoments where XenApp was used only to log the incoming user and start a Firefox on the farm. Well, that Firefox could be retargeted to any other internal system, or local file. Plus it could have been used to go online and grab your exploits or code or files from your online accounts. Publishing "open" things is not good. Always try and publish only the final destination apps. If this cannot be accomplished, you're doing something wrong.

2. Browser is not needed.

A typical XenApp installation will include a Web Interface. Now, do you need it at all?
Consider this - starting up a browser, typing and selecting one of bookmarks to navigate to your various Web Interfaces is a drag. Why is it still being used? Only for flexibility of user access, so you could try and access your site from any external of uncontrolled system. Now, what if you have internal, corporate systems that are only accessible to internal clients?
For corporate apps, you should be using Citrix Receiver deployed on client machines, and Citrix Service Site instead of Web Interface. This will improve user experience. Modern Receivers allow users to drag needed icons straight into their Start Menu. One won't need to muck around in a browser to be able to start a pubilshed app. And there's no annoying "bye byes" from Web Interface where you have to re-login due to session expiration.

3. Check your security requirements.

Some still think that Citrix XenApp is a sort of security tool. It is not.
Consider this - you may have 55 users sitting on the same host simultaneously, each user's outgoing sessions have the same source IP. If you have 5 different systems, that source IP has to have route and ACL permissions to go to any system. This means that on network level all 55 users have access to all 5 systems (even though they need not). See why I raised a point 1?
Now to logging. One of your users has managed to run a full internal network scan, discovered some vulnerabilities (sometimes by looking at internal web source code from the published Firefox) and successfully exploited some loopholes, causing service disruption or credetnial leakage). You discovered the issue few hours later, when the farm had [another] 15 users on it. How do you go about tracking suspicious activity to a personal account? In your network logs, it's same IP for all users. On the XenApp host, you have no logs telling you what user had which TCP sockets open at a given moment.
Compare this to VPN now, where everyone has unique IP addresses, hence network permissions and route tables.
One might say - well what about SmartAuditor, a software that can record all user's session as MP4 videos?
OK, so that guy spent 4 hours on legitimate work, then issued a stray command that crashed your system, then kept working for another 3 hours - how are you going to catch it? Watch all 7 hours of his (and other users') videos and following closely everything they were doing? Hardly productive, and certainly not indexable.


4. Secure Gateware deployment.

There are guides on Citrix Secure Gateway deployment. Please be sure to carefully consider what is more important to you. The two main features of CSG are PRE-AUTHENTICATION and WEB-SECURITY.
Pre-authentication means that before your user gets access to your Web Interface, it is authenticated by the CSG.
Web Security mean you're not exposing IIS to externals. CSG offers it's own, sanititzed and locked down environment.
Do not neglect these two, or you will be neglecting the CSG's value. For these reasons, CSG should be deployed in its own box, which is hardened. Do not install Web Interface or other components on the same bax, this will negate the CSG's value or protecting those components.

5. VPN is still a viable alternative.

As already mentioned before, sometimes VPN is a more traceable, transparent solution. You can regulates access on the network level. Different groups can have different ACLs in the network, because they could receive IP from different subnets, or they may receive different routing tables (less secure).
Many negative things about VPN come from the past era of IPsec/L2TP which had it's share of compatibility issues.
Welcome to the age of SSL VPNs. One does not need to go far - Windows 7 (or Vista) support SSL VPN out of box.
If you're a bigger shop, do check out DirectAccess feature in Windows 7/Windows Server 2008R2.

6. Microsoft RDS may fit your bill.

Microsoft has improved and revamped it's Terminal Services to become Remote Desktop Services (RDS).
The new service can give you most of what Citrix XenApp allows, at much more affordable price.
Consider performance, too - XenApp 6.0 is much slower at opening sessions than MS RDS.
Most of XenApp's value - Advanced Load Evaluators, Offline Published Programs etc are rarely used by SMB or even big Enterprises. Being another software, XenApp certainly takes some extra effort in both integration as well as in daily use. Microsoft's offering is certainly better integrated and starts up faster. It also does not require separate skills.
So check your requirements.

7. Some apps are born to be wild.

With Web 2.0 era, more and more applications are moving to the browser (contrary to what I said in point 2 - but that was about way of accessing XenApp). This means that apps are thin enough to be run on any client, and the protocol in between the client and the server is resilient to low-speed or low-quality lines. Lines are getting better every day, so you may need to consider not having any extra "gateways" or VPNs in between the client and the server.
They are all about legacy or insecure software anyway...

Thursday, July 14, 2011

LinkedIn and laziness

Would you agree that LinkedIn is not "thinking for users" as others (esp.Apple) do? Basic virtue of the human nature is laziness, and it's well understood by the most successful consumer companies and inventions.

Consider TV Remote Control. Consider iPhone. Facebook...

Now on LinkedIn, the most "active" function is "connecting to people".
It actually makes me think that very soon everyone will be connected to everyone, withing their area of expertise.
Is this the ultimate goal for LinkedIn and it's customers? I doubt it.

LinkedIn is not thinking hard enough for its users. And its users, being business-oriented, have little time to actively use LinkedIn effectively.
And this is why it's not as popular as Facebook.

LinkedIn's user want to upgrade jobs. How does LinkedIn help this goal?
It helps people to connect and then relies on their actions.
As many have found out, the only reliable part of web user is inaction (well, make it a single click).

It would help if LinkedIn figured out your future job aspirations and found those jobs, and suggested them on a "single-click menu".

It would help if LinkedIn figured what people might be useful to you, and mutually introduced them (not based only on past, think "future").

It would help if LinkedIn suggested you groups to join, based on your skills and activity, and introduced groups to people, and well as people to group moderators.

Now all it does is "keeps you linked in"...

Thursday, June 30, 2011

Office 365 arrived

I have watched Microsoft's Office 365 introduction for IT Professionals.

It was interesting until they revealed you still need to install 800M distro onto every computer where you intend to have access to your "online" documents. And keep that patched. Oh yes, and no Mac/Linux version.

Next.

(The only advantage is low entry fee, which may be interesting for small businesses. But then you will have Office-format documents, and as you grow to medium-size, it won't be easy to migrate away from that format, just ask others. That's when the paying spree begins.)

UH-OH, and don't forget the Patriot Act!

Wednesday, June 29, 2011

Apple's iPad 2

Went to Mayer, tried iPad 2.
Body shape is much better, and weighs a little bit less.

But - same so-so screen, and yes I want it even lighter.

Overall it's not a good buy at 579-989 AUD.
Holding on until iPad 3 with Retina Screen...