The post was initially written in English for ImpactX: A ‘Smart Stick’ Puts The Internet of Everything In Service To The Blind and translated to French: Une canne intelligente met l’Internet of Everything au service des malvoyants

I must admit I’m really thrilled by this project, it’s innovative, it’s tech and the team that runs it is awesome!

The post My first guest post… In the Huffington Post appeared first on Remi Philippe.

The post Sam Altman – CS183B appeared first on Remi Philippe.

Hopefully, it’s pretty straightforward on OpenBSD, let’s assume $if is the interface connected to the Public Network , 88.190.xxx.xxx is the IP address (also called FailOver IP) given by your provider and 88.190.17.1 is the gateway.

When installing OpenBSD you can configure the IP of the device as a /32 without any problems, but when it comes to configuring the gateway it won’t work. Why? Because OpenBSD is (rightly) expecting to have the gateway on the same subnet so we have to work our way around that.

In the installation process, after configuring the network you have an option to make extra configs at the shell prompt, that’s our way out!
First of all configure the IP address of the interface:

ifconfig em0 inet 88.190.xxx.xxx netmask 255.255.255.255

If you try to ping the gateway it will fail as it’s not on the same subnet, so we need to tell OpenBSD how to reach this IP, to do this we’ll configure OpenBSD so that it will ARP for that IP on it’s external interface:

route add -inet 88.190.17.1/32 -link -iface $if

Lets try to ping the gateway …

# ping 88.190.17.1
PING 88.190.17.1 (88.190.17.1): 56 data bytes
64 bytes from 88.190.17.1: icmp_seq=0 ttl=255 time=0.551 ms
64 bytes from 88.190.17.1: icmp_seq=1 ttl=255 time=0.418 ms
64 bytes from 88.190.17.1: icmp_seq=2 ttl=255 time=0.463 ms
64 bytes from 88.190.17.1: icmp_seq=3 ttl=255 time=0.483 ms
--- 88.190.17.1 ping statistics ---
4 packets transmitted, 4 packets received, 0.0% packet loss
round-trip min/avg/max/std-dev = 0.418/0.478/0.551/0.054 ms

… and it works!

We just need to setup the gateway and finish our setup.

Now we want to make sure the interface starts at boot time, to do this we’re just going to create the file /etc/hostname.$if with the following content:

inet 88.190.xxx.xxx 255.255.255.255
!route add -inet 88.190.17.1/32 -link -iface $if

and populate the file /etc/mygate with the IP of the gateway

88.190.17.1

Here we are, OpenBSD is running in the virtual world but connected to the real world!

The post IP configuration on OpenBSD online.net (aka Dedibox) appeared first on Remi Philippe.

Moving from development to 1.8 a few things changed in the configuration (not much), here is a quick update

Step 1: Pre-Requisites

On the sRTP side we still need the libraries (which can be found here) otherwise you will see …

checking for mandatory modules:  CRYPTO MYSQLCLIENT SRTP OPENSSL... fail

configure: ***
configure: *** The SRTP installation appears to be missing or broken.
configure: *** Either correct the installation, or run configure
configure: *** including --without-srtp.

Refer to the sRTP on Asterisk article for how to set this up.

If you plan to include MySQL support on debian, don’t forget to install the libmysqlclient-dev

toera-g-sip1:/sw/src/asterisk-1.8.1.1# apt-get install libmysqlclient15-dev
Reading package lists... Done
Building dependency tree
Reading state information... Done
The following NEW packages will be installed:
  libmysqlclient15-dev
0 upgraded, 1 newly installed, 0 to remove and 0 not upgraded.

Step 2: Configuring the binaries

First of all lets configure the binaries, here is the command I used

toera-g-sip1:/sw/src/asterisk-1.8.1.1# ./configure --prefix=/sw/asterisk-1.8.1.1 --with-crypto --with-ssl --with-mysqlclient --with-srtp

The options are self explanatory, –with-crypto, –with-ssl are for the TLS part, –with-srtp for the … sRTP part and finally –with mysqlclient is for the MySQL support (not required by sRTP or TLS, but used for CDR).

Once this is done, run make menuconfig to enable the TLS and sRTP modules (see the original article for details on that).

Step 3: Configuration part

Here are some changes, first the option in the sip.conf file is not anymore

srtpcapable=yes

but

encryption=yes

Second the dial plan part in the extensions.conf file is not

exten => 9999,1,Set(_SIP_SRTP_SDES=optional)

but

exten => 9999,1,Set(_SIP_SRTP_SDES=1)
exten => 9999,2,Set(_SIPSRTP=1)
exten => 9999,3,Set(_SIPSRTP_CRYPTO=enable)

And basically that’s the only changes there are compared to the development version. As usual you can check if it’s working through a sip set debug on :

<--- SIP read from TLS:85.xxx.xxx.xxx:55660 --->
INVITE sip:01xxxxxxxx@sip.remiphilippe.fr;transport=tls SIP/2.0
Via: SIP/2.0/TLS 192.168.0.17:46679;branch=z9hG4bK-d8754z-a51806220748b135-1---d8754z-;rport
Max-Forwards: 70
Contact: <sip:100@85.xxx.xxx.xxx:55660;transport=TLS>
To: "01xxxxxxxx"<sip:01xxxxxxxx@sip.remiphilippe.fr>
From: "Remi"<sip:100@sip.remiphilippe.fr>;tag=bd9bb73d
Call-ID: ZTE0ZTdiODBhYTA0MDQ2ZjYwNDZlMTAyYTM0Y2ViZmQ.
CSeq: 1 INVITE
Allow: INVITE, ACK, CANCEL, OPTIONS, BYE, REFER, NOTIFY, MESSAGE, SUBSCRIBE, INFO
Content-Type: application/sdp
Supported: replaces
User-Agent: Bria 3 release 3.1 stamp 58312
Content-Length: 593

v=0
o=- 1295179812598219 1 IN IP4 192.168.0.17
s=Counterpath Bria 3.1
c=IN IP4 192.168.0.17
t=0 0
a=ice-ufrag:4de55c
a=ice-pwd:d96683ec023786e73ca78e87d0383765
m=audio 53080 RTP/SAVP 9 0 8 18 101
a=rtpmap:18 G729/8000
a=fmtp:18 annexb=yes
a=rtpmap:101 telephone-event/8000
a=fmtp:101 0-15
a=crypto:1 AES_CM_128_HMAC_SHA1_80 inline:2aHmG4LMIzu51D93lMPZpr5HCPuSgsmDpgM4siEz
a=crypto:2 AES_CM_128_HMAC_SHA1_32 inline:2aHmG4LMIzu51D93lMPZpr5HCPuSgsmDpgM4siEz
a=sendrecv
a=candidate:1 1 UDP 659136 192.168.0.17 53080 typ host
a=candidate:1 2 UDP 659134 192.168.0.17 53081 typ host
<------------->

Here we are!

The post Asterisk sRTP with 1.8 appeared first on Remi Philippe.

Here is a brief history of my journey to the CCIE SP…

8 September 2009 – CCIE Service Provider Drake

Nothing special here, I decided to go for the CCIE Service Provider in July so I spent 2 months working the theoretical stuff based on the usual books (Routing TCP/IP Volume 1 & Volume 2, MPLS Fundamentals from Luc De Ghein, MPLS VPN and Architectures Volume 1 & Volume 2 mainly with lots of bits and pieces online) and then took the drake exam. It’s a normal CCIE Drake I actually even found it more easy than the R&S one I got something around the 90%.

14 December 2009 – First Attempt

My first attempt for the CCIE Service Provider lab, there were no Open Ended Questions (OEQ) at the time so didn’t have to study “core knowledge” even though through my preparation for the Drake I’d expect to have enough knowledge to get through.

The lab by itself wasn’t that hard, actually it was the easier one of the 3 labs (from my point of view) I really missed a chance to pass on first go. Why did I fail? I wasn’t prepared enough, I was in a period with a heavy work load and I only spent one week studying on evenings, way not enough compared to what I had done for the R&S lab. My knowledge of MPLS VPN was limited and I clearly didn’t have enough knowledge on Multicast in SP environment.

The score report was no surprise, it really reflected my lack of knowledge in those topics.

16 April 2010 – Second Attempt

This time I was much more prepared, I took one week off to work full time just before the exam and took 2 weeks of evening work before. The new thing was the OEQs I had to work “core knowledge” as well and as the drake was far away it was like starting from 0… That was mainly my first week, I spent lots of time working on that part. I spent my second week working on “mini labs” such as the labs of Antonio Soares and more specific labs like L2VPN, PPPoE, BGP tips and tricks etc … That’s actually when I started to blog the tech articles, instead of keeping a notepad with all my findings I decided to put them online. Finally the last week before the exam I did full length labs from IE to get in the 8h rhythm.

The error I did there was to take a lab on the Friday. Friday you’re tired it’s the end of the week and I really felt it, after lunch it was REALLY hard to get back into the lab.

Why did I fail there? A few reasons I think, the first one I was tired so I felt more stressed, much more than usual. The second one I hadn’t worked BGP enough this time (send-label for example) so I spent lots of time on stupid things with CCO.

13 July 2010 – Third Attempt

I learned from my mistakes in the first and second lab attempts. I really did a hell of a work on BGP and MPLS not with full length labs but with lots of mini labs where I really took lots of time to understand the debugs and outputs.

But the biggest work I did was my mental preparation, I was not stressed at all when I went for the lab, I felt totally prepared and it really changed everything.

In terms of strategy, I first went through the OEQs, they were quite doable I think, maybe some of them were a bit tricky but if you take the time to read them properly it should be ok. For the lab by itself I started by reading the whole lab rapidly, then did the tasks that would require a reboot and went back in the book while the routers were rebooting. I read everything and started with the IP Services / Security questions (those are easy points but they can break your lab if you’re not careful, by doing them first if something goes wrong you’ll see it before the end), where possible I also did the QoS questions. Once this was done I went through the rest of the lab in the normal order. I was finished at 3pm and had until 5pm, so I went around all the routers and tested everything. I found a few mistakes there that I corrected.

In the last minutes I did a few changes, and that was stupid as I wanted to be fast I forgot things after I did those changes so I lost a couple of points. Of course I noticed that AFTER my lab… DO NOT make changes in the last 30 minutes of your lab, bad idea.

I was comfortable I had done quite well, but because of this stupid change I thought that would screw up my lab… So I went with the idea I had failed…

Took the Thalys back home and went to bed around 9pm because I didn’t feel like waiting for the results. I woke up at 4am (thirsty not to go and check the results) so I decided to give a quick check to see if the results were there…

Haha, the email is there as usual no results in the email you have to go and check (and the website was slow, imagine how patient you can be at 4am …). Finally I logged in and saw the results … PASS !!! But actually I thought I was dreaming so I just left the laptop and went back to bed. 15min later I wondered if I had seen right so I woke up again went back to check and there I really shouted my joy

The end …

Finally this is the end of my certification track (for the CCIE at least), 2 CCIE are enough and I won’t be going for a third one it’s not useful and it’s too much work. I still have to post one or two tech articles on the CCIE prep and that will be the end. I’ll of course continue blogging on network / IT stuff and business!

The post CCIE Service Provider – Done! appeared first on Remi Philippe.

The idea was to edit the photo in order to have everything in black and white except Caroline and the computer screen.

Doesn’t look bad, though it’s a pity the bottle of water was in front on the screen…

The post Have you ever seen a webdesigner at work? appeared first on Remi Philippe.

There is a help link next to the promoted button, it seems they have introduced this on the 16th of June

Update: June 16, 2010
Q. What are “Promoted Trends”?
A. Promoted Trends are a new advertising concept we began testing this week; they are an extension of our Promoted Tweets platform. With Promoted Trends, users will see time-, context- and event-sensitive trends promoted by our advertising partners. These Promoted Trends initially appear at the bottom of the Trending Topics list on Twitter and are clearly marked “Promoted.” As conversations about the topic increase, Promoted Trends may move up the list.

Source: Twitter Support

Can’t wait to see if that’s going to take off!

The post Has Twitter started advertising? appeared first on Remi Philippe.

I couldn’t get mine on the 24th as I was busy so it arrived at Cisco for me and waited until Friday. On Friday got it, unpacked it, loaded the backup of my 3GS and put in the new Micro SIM card. I had to call the 0800 100 740 which is the Orange SIM card activation number, typed in my 13 digit number for the SIM card and … DRAMA… The number you typed is incorrect. Of course I tried typing it in a few times before deciding to call the customer support.

Nothing… The phone was ringing out in the blue. I gave it up and tried again around midday, got someone who told me it was normal the servers were down because they were too many activations. I kept trying regularly until it finally worked at 7:51pm.

Now Orange, really, how hard would it have been to post a message on you iPhone dedicated site? Was it really so though you couldn’t close the activation portal until the servers returned to a normal load and reopen it gradually? On your 4th iPhone launch couldn’t you anticipate this high demand? iPhone were only available through pre-orders, couldn’t you have anticipated a 50,000 subscribers trying to activate their SIMs?

Result of all this? Fantastic advertisement everywhere on how the iPhone screwed up Orange (yes because it also screwed up all other SIM card activation). Wouldn’t it have been simpler to just communicate instead of having all the blogs / tech sites / news site do it for you?

On my side that had a major impact on the quality I perceived from Orange (which was very high until today…). Lets see how you’ll handle communication to impacted customers now…

The post How Orange screwed up the iPhone launch appeared first on Remi Philippe.

A bit of history

February 2005

I got my first 3G phone a Nokia 6680. It was the first 3G / Video phone available by Orange at the time. At the time finding someone with a video capable phone was hard, but to try and launch this service Orange was offering free video calls on their network. That was the old times where no one hardly had a data plan on their phone so bandwidth utilization was not a concern.

Did the service take off at this point? Nope. Why? I think there was 3 factors.

The first one was technological, the screen is tiny, the quality of the video is quite poor so it wasn’t really interesting for the users.

The second one was the applications, why would you need it? Was their any killer device out there to make you want one at any cost? Was their any killer software around? No! The only interest would be to talk with your friends in video, but there why would you do it? Can’t you wait to get back home and use Skype where you’d have a bigger screen and higher quality video for free?

The final one, would be support! Why support? Who was really promoting the video over 3G? The mobile operators and some vendors. Did you see big adverts about this service? I didn’t the only adverts I saw were from the mobile operator or the phone manufacturers.

So what was the business model? It’s hard to tell but clearly they was no real decision from anyone to push the service… For manufacturers it didn’t really change much, most of the phones already had a camera and the video stack had been implemented in their software. For mobile operators it was more of a feature to attract people on their brand new 3G networks.

June 2010

This is where it gets interesting, at the WWDC (Apple’s developer conference) Steve Jobs introduced the video call functionality called FaceTime available only on iPhone4. How is it different from the previous services? First of all currently FaceTime only works over WiFi between iPhone4 (not sure if it works between carriers must test that but it should as it uses WiFi) and secondly … It’s Apple!

The “Its Apple” is the most interesting part of it all, the technology is not revolutionary at all, it’s probably derived from iChat and video calls exist since 2005. But what Apple is really good at is marketing. How many iPhones were sold on the launch date? Estimates say around 1.5 Million find me another device that did that well! Why is everyone (including me) running to get this new iPhone? It’s a phone, you can call with it, it has a camera like most phones, just like BlackBerry and Androids there is an AppStore.

There are a few things that make the success of Apple

– The Fan Boys, lots of buyers are really Apple fans, there are committed to Apple and can even accept problems they wouldn’t with other products (the signal issue for example). They are really captive customers and acquainted to Apple’s cause.

– The simplicity, it’s just so easy to use, update and so on… On the other hand a mobile phone is also very easy to use, but where can you get a PDA / entertainment / phone device that works out of the box, with tons of applications and very stable? Android? I doubt it’s more of a geek device where you can do almost anything. Blackberry? It’s more of a business device even though they are making huge efforts to add new functions in order to change it to a mobile entertainment device. The iPhone is to my sense the only device that can provide a work and play device (it’s even trying to compete with the PSP/DS!).

– The design, it’s just beautiful Apple has always invested lots of energy in the design making it the “hype” device. Geeks may love it for technology though it’s limited (for their use) but most people will love it for the design / functionality / fashion (with a strong + on fashion).

So where is the difference

They are quite a few differences with the previous model, the biggest one being the fact it’s driven by the manufacturer and not the mobile operators, but lets cover some of the differences

• The first one, the adoption remember, estimates say that Apple sold 1.5 million devices, that means 1.5 million potential FaceTime users and that number will grow.
• The second one, remember it’s simple and nice, the screen is bigger and quality is better.
• The third one, it doesn’t require the mobile operators to get new setups on their network because it will go through WiFi (remember what Steve said on that point, currently). I suppose the idea is to create a user base on WiFi to push phone operators to deploy the service for a fee.
• The forth one, video is taking a bigger importance in our lives think of Cisco TelePresence for example, you’re getting used to video conferencing
• The biggest one, guess who’s going to do the marketing? Apple will. And they are REALLY good at that. Remember they will be marketing their product to get more sales, they don’t really care about the mobile operator (to a certain extend) so if they really want it to take off they’ll do what is necessary.

With that I’m certain the video is going to take off, even though there are quite a few glitches at the moment (the quality is highly dependent on your network connection) but on another hand remember this is release 1!

The post Why will Apple succeed with video call service appeared first on Remi Philippe.

I was reading an article of Peter Bregman on the iPad at HBR (read it here). And I must admit I think he’s right.

The point that Peter is trying to make is that having an iPad can make you over-productive. This shouldn’t be bad right? But in fact it is not as good as it seems. When you don’t have anything to do you can take time to think, lets take an example.

I own an iPad (but I’m not ready to return it sorry Peter). I have it with me all the time, except this evening where I lent it to a colleague. That’s when Peter’s post made the more sense to me. Driving back home I hadn’t the radio on as I was on the phone this morning when I drove to the office and I was stuck in a traffic jam, usually what I do is just take my iPad and start wandering on the net while I’m stuck. What did I do instead? Reflect on my day, what I did compared to what I had to do, what I forgot (believe it, I used paper to write it down! :)), who I had to call the next day, …

The result? All those things I took time to think about didn’t start turning in my head as I arrived home. I had had the time to clear up my word day and when I arrived home, I really felt “home” not “word from home”.

Would it have been different if I had my iPad? Yes! Clearly, I would be playing around with it all evening until I went to bed and once in bed I would have started thinking about my workday and it would have bugged my sleep.

Would I change my iPad now? No! But I think I have to get a reasonable use of it (for example not bringing it in front of the TV). It’s still very useful on a workday and on weekends for photography or PERSONAL emails

The post iPad revolution or malediction? appeared first on Remi Philippe.