In een van mijn vorige (Engelstalige) posts ben ik ingegaan op het creëren van een online radiostation d.m.v. Radionomy. Zoals ik in die post aangaf is de lengte van te uploaden audio-items beperkt tot 15 minuten (iets korter zelfs). Eén van de problemen die ik ben tegengekomen, en met mij vele andere radiomakers bij Radionomy, is dat het door deze beperking lastig is om lange(re) DJ-mixen of vooraf opgenomen programma’s te gebruiken. Daarnaast stelt Radionomy nog de eis dat elk afzonderlijk nummer dat je draait vergezeld dient te gaan van de juiste artiest+titel in de mp3ID tag. Dit heeft te maken met eisen rond royalties maar lukt niet wanneer je een vooraf opgenomen programma of DJ-mix upload in stukken van 15 minuten. Voor zover ik weet, ondersteunt Radionomy namelijk geen “embedded” cue-sheets.

Ik heb echter een oplossing gevonden voor dit probleem. Deze oplossing maakt gebruik van een aantal tooltjes die het gemakkelijk maken om enerzijds een mp3 in stukken te knippen en anderzijds de geknipte mp3′s van de juiste mp3ID tag te voorzien. Tenslotte dien je tijdens het uploaden nog wat instellingen aan te passen in de Radio Manager, het programma dat elke radiomaker moet gebruiken om zijn of haar audiobestanden te uploaden naar de Radionomy servers. Ik ga er nu even van uit dat je Radio Manager al heb geïnstalleerd op je systeem en dat je bekend bent met de werking ervan.

Naast Radio Manager heb je dus nog een tweetal tools nodig: CueMaster [1] en mp3DirectCut [2]. Deze tools kun je vrij downloaden van het internet. Gebruik de links onderaan dit artikel om de tools te downloaden en zet ze op een handige plek neer (bijv. op je bureaublad).

Om de werkwijze uit te leggen, zal ik een praktijkvoorbeeld gebruiken. Voor de weekendprogrammering van mijn eigen radio station (Move! Dance Radio [3]) krijg ik van een aantal DJ’s de mixen aangeleverd als één enkel mp3 bestand. Dit bestand bevat dus de hele mix en niet als losse nummers. Daarnaast vraag ik van deze DJ’s ook een documentje of bestand waarin staat wat de playlist van de mix is en hoe lang de nummers duren. Deze informatie is sowieso nodig om te voldoen aan de regels aangaande royalties, namelijk dat je voor elk nummer de artiest en de titel moet meesturen.

Stap 1 – Cue-sheet maken

Om de procedure zoveel mogelijk te vergemakkelijken, maak ik van dit document een cue-sheet [4]. Daarvoor gebruik ik CueMaster. Start CueMaster op door te dubbelklikken op het icoon. Het startscherm van CueMaster ziet er zo uit:

Voor de geselecteerde mix kun je dan alle relevante informatie invullen in de velden die je ziet. Je kunt een nieuw item (=regel) invoeren door de toets Insert in te drukken. Je krijgt dan een kleine popup waar je drie velden kunt invullen: artiest, titel en tijd. De tijd heeft als format mm:ss:ff, waarbij mm staat voor minuten, ss voor seconden en ff voor frames. Ik laat de frames altijd maar voor wat het is, dus daar vul ik 00 in. Daarnaast kun je nog wat algemene velden invullen voor je audiobestand. Die velden spreken voor zichzelf.

Wanneer je klaar bent met invullen, dan ziet het scherm er ongeveer als volgt uit:

Je ziet dat ik 13 tracks heb toegevoegd (eentje valt net buiten het venster) en dat ik voor elke track de juiste lengte heb ingevuld. De totale lengte van de mix is 59 minuten en 47 seconden. Dit kun je eventueel checken door naar de lengte van het mix-bestand te kijken. Je kunt door de afronding enigszins afwijkende lengtes zien, maar dat is niet zo’n probleem.

Nu je alle informatie hebt ingevuld voor deze mix, kun je een cue-sheet exporteren. Klik hiervoor op File | Save As. Je krijgt dan een pop up menu waar je de file naam kunt kiezen. Het is verstandig, eigenlijk zelfs belangrijk, dat je de naam kiest van het mp3-bestand waar het bij hoort, maar dan met extensie ‘.cue‘. De cue-sheet bevat namelijk een veld waarin de naam van de bijbehorende mp3 wordt genoemd. Als deze naam  niet overeen komt met het mp3 bestand, heb je eigenlijk helemaal niets aan de cue-sheet.

De cue-sheet ziet er ongeveer zo uit:

TITLE "Floating Beats DJ Sessions 6"
PERFORMER "DJ Joshua"
REM ** Generated by CueMaster v1.6 **
REM Time  : 59:47:00
REM Length: 3587 seconds
REM Year  : 2010
REM Genre : Dance
REM DiscID: b00e010d
FILE "DJ Joshua - Floating Beats Sessions on Move! Dance Radio #6.mp3" MP3
..TRACK 01 AUDIO
....TITLE "Reel (Original Mix)"
....PERFORMER "Rene Breitbarth"
....INDEX 01 00:00:00
..TRACK 02 AUDIO
....TITLE "Formula Juan (Original Mix)"
....PERFORMER "Audiofly"
....INDEX 01 04:11:00
(...)


Zoals je ziet worden de tijden (=lengte van audio item) die je in CueMaster hebt ingevoerd omgerekend naar de startpositie in de file. De eerste track begint dus op 00:00:00 en de tweede op 04:11:00. De lengte van de eerste track is dus 4 minuten en 11 seconden.

Mocht je een PREGAP tag in de file vinden, dan kun je die m.i. maar beter weghalen, omdat ik niet weet hoe mp3DirectCut daarmee omgaan bij het splitsen van mp3 files.

Stap 2 – De mp3 splitsen

Nu we een cue-sheet hebben gemaakt, kunnen we CueMaster afsluiten. We starten nu mp3DirectCut op door op het icoontje te dubbelklikken. Het startscherm van mp3DirectCut ziet er zo uit:

Naar mijn bescheiden mening heeft dit programma een draak van een interface die helemaal niet intuïtief werkt. Gelukkig hebben we van deze interface vrijwel niets nodig omdat we in de voorgaande stap een cue-sheet hebben gemaakt!

We laden allereerst de cue-sheet in mp3DirectCut door in het menu File | Open te selecteren. We kiezen dan het bestand dat we in de vorige stap met CueMaster hebben gemaakt. Het resultaat van deze actie ziet er als volgt uit:

We zien nu de titel van het bestand en een versimpelde wave-form van het audiobestand, zij het slechts gedeeltelijk. Daaronder zien we een selectiebalk die is onderverdeeld in 13 blokjes. Dat is de balk die aangeeft waar in het bestand de cuts zullen plaatsvinden op basis van de geladen cue-sheet. Je weet vast nog wel dat ons bestand uit 13 tracks bestond.

We kunnen nu de mp3 file heel gemakkelijk splitsen door in het menu File | Save Split te selecteren:



In deze pop up kunnen we dan een folder opgeven waar we de gesplitste bestanden willen opslaan. Daarnaast kunnen we zelf een formaat kiezen voor de filenamen voor de bestanden. Ik kies hier altijd voor ‘%N %A – %T‘ zodat de bestanden altijd in de juiste volgorde staan (%N = track nummer). Wanneer je alle gegevens naar eigen smaak hebt ingevuld kun je gaan splitsen. Je hoeft hiervoor alleen maar op ‘OK‘ te klikken, het programma doet de rest.

Stap 3 – De bestanden uploaden

De bestanden kunnen we, zoals gezegd, uploaden d.m.v. de Radio Manager. Ik ga er nogmaals van uit dat je weet hoe dit programma werkt en hoe je audiobestanden moet uploaden. Start de Radio Manager, klik op het menu item Upload bestanden en klik op Persoonlijk item uploaden. Kies als item Muziek en selecteer daarna het eerste van de losse bestanden die je in stap 2 hebt gemaakt. Je zult zien dat de titel en artiest al correct zijn ingevuld. Dat hoef je dus al niet meer met de hand te doen. Kies nu Volgende. Je zult het volgende (inmiddels bekende) scherm zien:

Je ziet hier een aantal aanwijzers die je normaal gesproken gebruikt om fade in en fade out tijden in te stellen. In bovenstaand plaatje staan alle aanwijzers op de ´default´ waardes. Om er nu voor te zorgen dat alle losse bestanden die samen de mix vormen netjes achter elkaar worden afgespeeld, dien je alle aanwijzers zodanig in te stellen dat er geen fade in en fade out worden gebruikt en dat alle nummers van begin tot eind worden afgespeeld. Daartoe zet je de Fade In, Intro en Mix aanwijzers allemaal op 0’00’000, zet je Outro op maximaal en zet je Fade Out ook op 0’00’000. (Als je meer wilt weten over de exacte betekenis van de aanwijzers, dan verwijs ik je naar de handleiding van Radio Manager.) Klik nu op Voltooien en het uploaden zal starten.

Deze procedure herhaal je tot je alle nummers van de mix hebt geupload. Overigens, bij het eerste audiobestand kun je eventueel wél een fade in gebruiken en bij de laatste wél een fade out, maar bij alle andere bestanden moet je er op letten dat je de aanwijzers instelt zoals hierboven beschreven.

Stap 4 – De afspeellijst maken

Nu we alle bestanden van de mix hebben geupload, dan kunnen we een afspeellijst maken waarin de nummers netjes achter elkaar worden gezet. Klik hiervoor op Klokken en afspeellijsten en maak een nieuwe afspeellijst aan (je kunt eventueel ook een bestaande afspeellijst gebruiken). Je kunt nu alle bestanden die je zojuist hebt geupload in de juiste volgorde in de afspeellijst plaatsen. Als je dat hebt gedaan, dan ziet het venster van de afspeellijst er zo uit:

(Trucje: wanneer je alle bestanden in de juiste volgorde hebt geupload, dan staan ze in Radio Manager ook in de juiste volgorde. Selecteer ze allemaal en sleep ze naar het onderste venster.)

De afspeellijst in dit voorbeeld begint met een jingle (de uuropener) en daarna volgen alle muziekbestanden die tesamen de mix vormen. Omdat ik de fade ins en fade outs allemaal hebt onderdrukt, zullen de bestanden door de Radionomy server naadloos achter elkaar afgespeeld worden. Het klinkt dus alsof ik een ononderbroken programma of mix uitzendt, terwijl ik in werkelijkheid losse nummers laat afspelen. En omdat elk nummer de juiste gegevens bevat, namelijk artiest en titel, voldoe ik ook aan de regels die door Radionomy zijn gesteld.

Voor het zover is, moet je de afspeellijst nog ergens in je planning zetten. Ik ga er van uit dat je weet hoe dat moet, dus daar ga ik in dit artikel verder niet op in. Ik wens je veel plezier en vele luisteraars toe bij het radiomaken!

Links:

[1] http://cuemaster.org/

[2] http://mpesch3.de1.cc/mp3dc.html

[3] http://www.movedanceradio.com/

[4] http://nl.wikipedia.org/wiki/Cue_(bestand)

Today, I decided I want to install OpenCV (Open Source Computer Vision) on my Fox Board G20 to prepare for a possible project that I might be doing in the near future. Already having an OpenCV demo on a regular PC, I thought it would be worth the effort to try and get that demo running on my G20 as a true embedded application. I like to share my results with you, dear reader, in case you’d like to create an embedded OpenCV-based vision solution as well.

If you want to install OpenCV on your Fox Board G20, make sure you read this post first: How can I add memory to my Fox Board G20?

The default OpenCV library that you get when you apt-get install libcv-dev is version 1.0. That’s a bit old, wouldn’t you agree? For a demo project, we are using OpenCV 2.1.0, which is what I want to install [1].

Before I continue, first a little warning: this is a rather lengthy process. Not that you have to be at your PC at all times, but you’d better be prepared that you can’t use your Fox Board G20 for quite some hours.

Well, let’s get started and download OpenCV:

debarm:~# wget http://sourceforge.net/projects/opencvlibrary/files/opencv-unix/2.1/OpenCV-2.1.0.tar.bz2/download

In case you haven’t installed a GNU compiler, I suggest you do so now:

debarm:~# apt-get install gcc g++

There are some tools that you need to install and some that you may want to install. I suggest to install them all:

debarm:~# apt-get install bzip2 pkg-config

debarm:~# tar jxvf OpenCV-2.1.0.tar.bz2

After unpacking OpenCV, enter the directory OpenCV-2.1.0 that was just created, create a build directory called ‘release‘ and enter it:

debarm:~# cd OpenCV-2.1.0
debarm:~/OpenCV-2.1.0# mkdir release
debarm:~/OpenCV-2.1.0# cd release

One of the tools that you will need to build OpenCV is cmake. You can simply install it using apt-get:

debarm:~/OpenCV-2.1.0/release# apt-get install cmake

Now, we can start building the OpenCV build environment:

debarm:~/OpenCV-2.1.0/release# cmake -D CMAKE_BUILD_TYPE=RELEASE -D CMAKE_INSTALL_PREFIX=/usr/local ..

This cmake command will create a Makefile for the release target, which means you are ready for installing OpenCV:

debarm:~/OpenCV-2.1.0/release# make

You will see many lines similar to this one:

[ 1%] Building C object 3rdparty/lapack/CMakeFiles/opencv_lapack.dir/dcopy.o

and you’ll have plenty of time drink coffee, read a book or do some actual work. So make sure you have a full pot of coffee available (in any of these cases). Uhm, perhaps I should have mentioned this earlier… Still, you’ll have more than enough time to make a pot of coffee, drink it and make another pot of coffee.

Better still, you can start your build in the evening and have a good night’s sleep while your Fox G20 is doing all the hard work!

When, finally, your build has finished, you can install the libraries, include files and documentation:

debarm:~/OpenCV-2.1.0/release# make install

There you go! You’ve successfully installed OpenCV 2.1.0 on your Fox Board G20! You are now ready for taking it for a spin.

If you’ve installed pkg-config, you can easily find the include files and libraries for OpenCV:

debarm:~/OpenCV-2.1.0/release# pkg-config opencv --cflags
debarm:~/OpenCV-2.1.0/release# pkg-config opencv --libs

which you can also incorporate in your Makefile. But I’ll leave this as an excercise for the reader.

After having gone through this myself, I thought perhaps I should spare you this ordeal (did you see the spoiler at the beginning of my article?). That’s why I created a package for you to download and install in case you want to play around with OpenCV right away. Please beware that the package only contains the include files and libraries. Here it is:

http://www.asksander.com/download/opencv-2.1.0.tar.bz2

Just download the file onto your Fox G20 and install using tar (from within the root of your file system):

debarm:/# wget http://www.asksander.com/download/opencv-2.1.0.tar.bz2
debarm:/# tar jxvf opencv-2.1.0.tar.bz2

It will install the include files in /usr/local/include/opencv and the libraries in /usr/local/lib.

Have fun playing around with OpenCV! If at some point in the future you’ve created a cool feature with OpenCV on your Fox G20, let me know! I’d be very interested to hear about it!

Kudos

[1] OpenCV Installation Guide

I have to admit, this question doesn’t quite fit the bill. However, If you’ve ever tried to compile a large project on your Fox Board G20, you may have noticed that it just won’t work due to memory problems. At some point, you will see the mmcqd process run overtime and your system grinds to a halt. I recently ran into this problem while trying to compile some big project and remembered that I had exactly the same problem when I tried to update the Perl installation on my G20. Believe me, you do not want to wait for this situation to resolve itself: it just won’t. In my case, I’d actually given up on trying to compile my project.

Thankfully, I asked about this problem in the G20 forum in Google Groups and I got an excellent pointer from Jan Klusacek:

Hi,
if RAM is issue then you should try use swap. You don’t have to create
new partition, 100MB file should be enough. After create, you can use
swapon to activate it.

I figured I could explain his suggested procedure here in more detail.

The reason you run into trouble building such a project actually is the lack of sufficient RAM. However, you cannot (easily) add more RAM to the G20, so you need a different solution. The solution is: swap space. Since kernel 2.6 you can use swap files with Linux; before that, you could only use a swap partition. So, we simply need to create a swap file on the G20. We don’t really need a lot of space, by adding a swap file of 100MB, you’re effectively more than doubling your memory, so this will suffice.

First, we need to create a swap file [1]:

debarm~# dd if=/dev/zero of=/myswapfile bs=1024 count=102400

This will create a file of 100 x 1024 blocks each having a size of 1024 bytes: 100 x 1024 x 1024 = 100 MB.

Next, we set up the swap space:

debarm~# mkswap /myswapfile

and immediately turn it on:

debarm~# swapon /myswapfile

Checking whether the swap is there and whether it’s used can be done using free:

debarm~# free -m
...... total     used     free     shared     buffers     cached
Mem:      59       54        4          0           3         41
-/+ buffers/cache:  9       50
Swap:     99        6       93

The option -m tells free to show you the results in Megabytes, which makes for a more readable result for us humans.

Finally, you probably want to use your swap space after a reboot as well. Use your favorite editor (I use joe, simply apt-get install joe to get it) to open /etc/fstab:

debarm~# joe /etc/fstab

and add the following line to it:

/myswapfile swap swap defaults 0 0

in order to automatically start using your swap space after a reboot.

That’s all! No wait, there is something else I have to say about it. Be prepared that this is in no way a fast solution. Memory reads and writes to RAM are many, many, times faster than memory accesses to flash memory. And since the root file system of the G20 resides on an SD memory card, flash memory is what you’re dealing with. So it’s up to you to decide whether this trade off is acceptable to you.

I finally got my project to compile, even though it took the better part of a day to get there, so I’m more than happy with this solution!

Kudos

[1] Linux add a swap file howto

In my previous article, I talked about Live365 as a platform for creating your own radio station. In this article, I’ll discuss the Radionomy platform that you could also use to have a radio station!

Radionomy

The other option for creating your own radio station, at least, that I know of, is Radionomy (www.radionomy.com). I have had actual experience with Radionomy for some time now, since my own radio station Move! Dance Radio is hosted by them.

Let me begin by saying that you don’t have to pay to be a broadcaster with Radionomy. Okay, there’s more to it, but you never have to pay Radionomy anything for using their broadcasting platform. However, in some cases you might be responsible for paying royalties. I’ll get back to this later in this article.

In order to create your radio station, you need to sign up with Radionomy. This is the easy part. Just go to the site, and click ‘Create’. The rest will basically show itself. Please make sure that you pick your station name carefully, you cannot change it afterwards!

After signing up and choosing a name for your station, you need to download the ‘Radio Manager’. This is an application that you will use to create playlists, clocks, or generally speaking, your radio planning. You will also use Radio Manager to upload your songs, in case you’re not using Radionomy’s own audio libraries. I don’t think there is an actual limit to the amount of audio that you can upload.

When you start uploading your songs, here’s what you’ll be running into. First of all, audio clips (songs) can never be longer than 15 minutes. I’m not quite sure what the idea is behing this limitation, but that’s just the way it is. You can try to upload songs longer than 15 minutes, but Radio Manager simply won’t allow you to do so. I myself find this a rather annoying limitation. I have some DJ’s providing my station with fantastically sounding DJ mixes, which normally have a duration of one hour or more. I have to cut them down to <15 minutes parts, which naturally breaks the flow…

For each song that you upload, there are a number of parameters that you set. First of all, you must fill in the artist and the song title. I guess this is necessary for keeping track of royalties.

This article is not yet finished…

If you’re a bit like me, and love radio, there’s a chance you might want to create your own radio station! There are a couple of options you may explore in order to set up your own radio station. Two options I know of are Radionomy and Live365. In this article, I’ll focus on Live365.

Live365

First of all, there’s Live365 (www.live365.com). I myself have hardly any experience with Live365, and my little experience is also outdated, so I can only give you the bare facts. I will focus on the Personal Broadcasting Packages. There are also PRO packages available for professional broadcasters, with completely different conditions.

Live365 is not free. The cheapest plan is US$5,95 a month, which is called the Intro Level, and for that money you’ll get 200MB of storage and a maximum of 6 listeners (3 Basic and 3 Live). The listening hours per month are limited to 2,232 for this plan, which is derived from the maximum number of simultaneous Basic listeners: 31 x 24 x 3 = 2,232.

A storage size of 200MB translates to a total play duration of 7 hours at 64 kbit/s, but for better sounds quality you should use 128kbit/s, in which case the stored audio is only 3.5 hours long. Or, put another way, you can upload approx. 62 songs each lasting on average 3.5 minutes. From my experience, this is simply not enough to offer a somewhat varied playlist to your listening audience, unless you’re willing to upload different songs every day!

Live365 offers the ability to use MP3Pro, which can be used at 64kbit/s with a sound quality that’s allegedly as good as a normal MP3 stream at 128kbit/s. This way, you could squeeze approx. 125 songs in your catalog, still based on an average duration of 3.5 minutes per song.

To give you an idea, radio stations with a very narrow focus have some 250-500 songs in their catalog. Other stations have 1,000+ songs in their catalog. This gives you an idea of the number of songs that are played, the actual number of songs in their music library might be much bigger than that.

There is no limit to the actual playlist length, you can repeat any and all stored audio clips as often as you like. “Real” radio stations play popular songs (current hits) much more often than other songs, don’t they? You can do the same, although this greatly depends on the type of station you wish to create. I don’t think you can pull this off if you’re operating a ‘golden oldies’ station.

You DON’T have to pay any royalties yourself, Live365 does that for you. This is important information, so I’ll repeat it:

You DON’T have to pay any royalties yourself, Live365 does that for you.

Live365 offers several broadcast packages. The cheapest, as I already mentioned, is Intro. The other levels are P1 (US$9.95 /month) through P6 (US$99.95 /month). Each higher level gives you more storage space (up to 6GB), more simultaneous listeners (up to 160) and therefore more listening hours per month (up to 74,400).

Please note that VIP members have unlimited access to all streams and are not counted in either the number of simultaneous listeners or the listening hours per month.

You can actually earn cash by being a broadcaster with Live365. There are two ways to earn money: with One-time Bounties and with the Subscription Share Program. The One-time Bounty program gives you a cut of 25% of the initial sale when “a listener subscribes to VIP in order to gain access to — or show support for –” your station. The Subscription Share  pays rewards based on the amount of time VIP members spend listening to your station in a given month. Live365 states that broadcasters can compensate their monthly fee by selling only a couple of VIP memberships.

If you’ve read this article and have recent and hopefully more profound experience with Live365, please let me know (just send in a comment). I’d love to hear from you, also from a personal interest in radio broadcasting.

When I started hosting one of my other sites (Move! Dance Radio, to be exact), I was given free Facebook advertising credit worth US$50. I decided that now would be a good time to see if I could use Facebook ads to generate publicity for both this site and the one I just mentioned. So, I signed myself up for Facebook ads, which meant filling in your credit card details even though the credits are free. I understand that, Facebook hopes that you’ll keep advertising after you’ve ran out of free credits. We’ll see about that!

When I started my campaign on Facebook, I ran into a number of acronyms that I knew, sort of, but didn’t quite understand. We’ve all seen acronyms like CPM, CTR and CPC in online advertising, but what do those actually mean? And how does it help you with your campaign? But let’s first have a look at those acronyms.

CPM – Cost Per Mille

CPM is the price you pay for 1,000 (one thousand) showings of your ad. Let’s be clear here: it does not mean Cost per Million, although I do understand the confusion. You can also remember this knowing that the roman number M is actually 1,000. Sometimes, the acronym CPT (Cost per Thousand) is used.

CPC – Cost Per Click

When someone sees your ad, decides to click on it and is therefore redirected to your web site, that click is charged to you. The amount of money that is charged for the click is the CPC.

CTR – Click Through Rate

CTR is the relation between the amount of clicks on your ad and the number of times your ad was shown. Put into a simple formula, it would look something like this:

CTR = (number of clicks / number of impressions) x 100%

I believe it is always shown as a percentage. At least with Facebook, it is.

The first parameter is hard to quantify, because it depends on the targeted result of your campaign. Do you want as many people as possible to see your ad, or do you want as many people seeing your ad to click through to your web site?

Basically, when you want as many people as possible to see your ad, you should choose CPM. You just want to show your ad, but are not interested in traffic to your web site, perhaps because you just want to create brand awareness.

When you want as many people as possible to visit your site and you want to control the amount of money you pay for your campaign, you should choose CPC. You really don’t care how many times your ad is shown, but you want to make sure you don’t start paying crazy sums of money when a lot of people (relatively speaking) actually click through to your site.

But is it really that simple? I don’t think so. That’s where the second parameter, the CTR, comes into play. I guess you were already wondering where that would come in! Let’s go through some examples to see how everything adds up. For both examples, we’ll use the following numbers (based on my Facebook experience so far, these numbers could be real):

CPM = US$0.45

CPC = US$0.15

Page views = 100,000

Example 1

Let’s assume we have a CTR of 0.01%, which means that only 10 ads have been clicked on. Now we can calculate the costs for payment based on CPM and CPC:

CPM-based: total cost = (100,000 / 1,000) x US$0.45 = US$45

CPC-based: total cost = 100,000 x 0.01% x US$0.15 = US$1.50

Example 2

In this case we assume a CTR of 0.45%, meaning that 450 ads have been clicked on! Once again, we can calculate the costs involved for both CPM and CPC based payments:

CPM-based: total cost = (100,000 / 1,000) x US$0.45 = US$45 (same amount)

CPC-based: total cost = 100,000 x 0,45% x US$0,15 = US$67.50 (way more)

There you have it. It’s really important to know the CTR for your campaign before you can make a sane decision between payments based on CPM and payments based on CPC. Perhaps you should even do a small ‘pilot’ campaign to find out what the CTR for your specific ad is before launching your campaign for real.

There really is a difference in CTR for different ads! Looking at my own campaign, I have three ads running. The ad with the highest CTR has a click-through rate that’s more than 4 times that of the ad with the lowest CTR! In my example, the difference was bigger, but that was purely to make a point.

In my previous posts, I answered questions about 1) how to share documents across several locations, and 2) how to send a large file or large files to someone else. Today, I received a tweet (yes, apparently my @asksander tweets are actually found) that allows me to answer the question of this post: how can you share files with someone else?

I was referred to FileShareHQ (http://www.filesharehq.com). It has a few service levels, or plans, of which only the most basic one is free. Other service levels offer more space (up to 99GB), more users (up to 100) or ‘Branding’, which, I’m guessing, allows you to change the look & feel from FileShareHQ’s to your own. This is only available to business users. Prices for the service levels range from US$5/month for the Personal Plan with 5GB of storage and up to 5 users to US$49/month for the Business Plan with 100GB of storage and up to 100 users. I am a bit confused about this, because on their home page they advertise with up to 99GB. I know, the difference is small, but these things make me wonder.

Update: it is possible to upload a single file of 99GB, while the total storage space is 100GB. [Source: @FileShareHQ]

The Basic Plan offers up to 1GB of storage and up to 2 users and is free, as I already mentioned. You do need to sign up before you can use the service, but that’s rather easy. Simply fill in your name, a user name, a name for your upload location, your email address and a password. After signing up, you will receive an email in your inbox welcoming you to FileShareHQ.

After logging in to FileShareHQ, you will automatically go to your Home directory. There you will first be presented with the credentials for using the FTP access of your account! Yes, you can access your files using an FTP client of your choice. Please note that you don’t need to use an FTP client, you can also upload files using your browser. I tried uploading a file using WinSCP, and that worked like a charm. I think there is something to say for using an FTP client, because in my experience the method of using the ‘Open File’ dialog of your browser is sometimes a bit cumbersome.

Adding folders is just as easy using the browser as it is using an FTP client. The same goes for removing folders. Or files, for that matter. In fact, there seems to be no difference in functionality, so it’s up to you to decide which method you prefer.

It is also very easy to add users. Naturally, you can not do this using an FTP client, you need to do this through the web site. Simply click the tab ‘Users’ where you will be shown the available users. At first, there won’t be any, but this is also the place you need to go to add a user. Simply click the button ‘Create a new user’ and you will be presented with a form where you can fill in the new user’s details. You need to fill a name, a username (which is automatically prepended with “<your-username>_”), his or her email address and a password. Finally, you need to decide on the user’s home folder.

While adding users, I was a little surprised. Remember, the Basic Plan offers up to 2 users. Apparently, I myself, as the account holder, wasn’t counted as a user, even though I can access all files! So, if you count yourself as a user, which in my opinion is fair, you can have up to 3 users with the Basic Plan! When you’ve reached the user limit, you will be informed about it with a warning box that says “You’ve almost used up all your users!”. I find that funny, since I had already used up all of my 2 (extra) users. I would expect a warning like this to show up when you’ve only got 1 user left. This is not an issue of importance, though.

The users you have added will be sent an email informing them that an account was created for them at FileShareHQ. It provides them with their user names and passwords along with the URL of their home folder. Using their personal credentials, they can log in to their accounts and start adding or deleting files or download existing files.

The site looks good and is easy to work with. Signing up is a breeze and I really think that offering FTP access is a nice feature. Perhaps in the future they will also offer Windows Explorer and (Linux) Nautilus integration on the desktop. That, and just a bit more storage for the Basic Plan, would make this an even more mature product and a real competitor to Dropbox.

Perhaps, in the future, I might do a comparative article about FileShareHQ and Dropbox. If you’ve read this article and would like me to do so, just let me know. You know where to find me.

Sometimes you have to send a rather large file to someone you know. Think files sizes of 10 Mbytes or more. Files this size cannot be sent as attachments to emails, since most email servers decline emails with attachments this large. Perhaps Gmail allows it, but not everyone in the world uses Gmail, and this will always be the case.

Option one: YouSendIt

One option to send a large file is to use the online service YouSendIt (http://www.yousendit.com). The basic and free service that YouSendIt offers is to send one file to someone else. I believe you can send it to multiple recipients at once. The limitation is that it allows a file size of max. 100 Mbytes, but in many cases this is sufficient. And it’s really simple to use as well, even though you need to sign up even for the basic service.

After signing up and logging in to the web site, you will be directed to the ‘Send’ page. On that page, you can enter the email addresses of the recipients and it even offers a rudimentary address book for storing your contacts. Optionally, you can enter a Subject and an accompanying message. Finally, you can add  the file you want to send. Remember, the file cannot be larger than 100 Mbytes.

There are a couple of options you can additionally choose, but those are not free. Prices vary from US$3.99 for a service like ‘Return Receipt’ to US$8.99 for Premium Delivery. The latter option allows you to send files up to 2 GB with basic tracking, 14-day expiration, and unlimited downloads.

When you click the ‘Send’ button, the file will be uploaded to the YouSendIt servers, which may take some time depending on the speed of your uplink. When the upload has finished, the recipient(s) will be notified by email that there is a file available for them to download. The download link will be valid for seven days, after which the file will be deleted.

If you want to send more than one file, you need to pack them together in a single file using one of many packing tools like zip, rar, tar or 7-zip. Just use an often-used packing tool so you can be sure the recipient can unpack the file!

Option two: WeTransfer

Another option for sending large files is WeTransfer (http://www.wetransfer.com). WeTransfer looks even easier to use than YouSendIt and allows you to send files up to 2 GB. The main page offers an amazingly simple interface to send one or more files. You can keep adding files until you’ve reached the 2 GB limit.

The next field allows you to enter up to 20 email addresses! That should be sufficient for almost all situations. Next, you enter your own email address and an optional message and click ‘Transfer’. The file(s) will be uploaded and a notification email sent to the recipient(s), which you entered on the main page.

WeTransfer informs you that the background image of their web page is actually an advertisement. This allows the people at WeTransfer “to buy coffee and to keep our servers running”. If you want the service to remain free, you might want to consider clicking on an advertisement that suits you. By the way, I applaud their honesty and openness about the advertisement.

The recipient(s) will be sent a notification mail to inform them that there is a file waiting to be downloaded. The link and the file will be available for 2 weeks. After clicking the link, the recipient will be sent to a WeTransfer web page showing a download button and…. an advertisement.

Conclusion

I have used both YouSendIt and WeTransfer for sending files. All in all, I believe that WeTransfer is the easier one to use. Also, you don’t need to register, which is a drawback of YouSendIt. And even though the basic service of YouSendIt is free, additional options are not. Some of these options, like file sizes of up to 2 GB and a storage time of two weeks, are standard with WeTransfer. The only drawback of WeTransfer is the ‘in your face’ advertisement that’s presented on the main page for both sender and recipient.

So there you have it. You might want to check out each of these great services to send large files to friends or other contacts in order to find out which of these services suits your needs the best.

Today, I ran into an interesting problem. Granted, it was not really my problem but someone else’s, but it made me curious about a solution. The problem was brought to my attention via Twitter. This guy apparently saves images at work all day using the ‘Save Image As’ feature of Firefox. He must have quite a different job than I do, because I don’t ‘Save As’ images on a regular basis.

He was annoyed about the fact that Google Chrome does not support ‘Save Image As’. I was surprised and immediately tried to do it with a random image on a random site that I had just opened in Chrome. I right-clicked on the image and voilà! The very first menu item is ‘Save Image As’. So, this doesn’t look like an actual problem at all.

Then I realized that saving background images is a different story. Just try it: if you right-click on the big @ sign on the home page of this site, you will see quite a different context menu. You will quickly notice that there is no ‘Save Image As’ item in it! I really don’t know why Google doesn’t want us to save background images, but there you have it.

If you’re the person who doesn’t like to be told by Google (or anyone for that matter) what you can and cannot do and you still want to save the (background) image, here’s how you can do it. If you look at the context menu after right-clicking the background image, you will see an item ‘Inspect Element’, or something similar, because I use a Dutch Chrome. If you click it, the Chrome Developer Tools window will open showing you the elements of the page. The element you clicked on will be selected. If you clicked on the image on my home page, the element selected will be ‘heading’.

On the right side of the window, you will see a menu bar. Most probably, its section ‘Styles’ will be opened. In it, you will see an entry ‘#heading’ and that shows you all sorts of information about the attributes that make up the style. And, you will see a mention of the background image! The good part is, that the actual URL of the image is mentioned there.

Having gotten this far, it’s really easy to save the background image. Just right click the image’s URL and the menu will give you a few options, including ‘Save Link As’. Just select that and you’ll be presented a nice ‘File Save’ dialog, which gives you the options to save the image anywhere you like and with any name you like!

I haven’t dared to suggest this approach to my Twitter friend who brought up the problem. And I do admit: it is a rather lengthy procedure for something simple as saving an image, even when it’s a background image. So, if anyone from Google reads this, please, PLEASE, add the option ‘Save Image As’ in the context menu for background images. We thank you!

We all know the problem. You’ve downloaded a file at home, where you have a huge bit pipe connected directly to the internet, that you’d also like to be able to access from your work place. Perhaps the bit pipe there isn’t quite as good, or you forgot the URL where to download the file. In that case, you would need something to share the file between work and home. You could email the file to yourself using Gmail, but for some reason that just doesn’t feel right, does it?

My suggestion would be to use Dropbox. It’s a great sharing application that works on several platforms. So, you can use it from home and from work and you can use it on a Windows machine, a Linux machine or a Mac! Apparently, you can even use it on your iPhone. However, an iPhone is generally carried around (and showed off) all the time, so I don’t see the point of installing Dropbox on that. You could just put your files on it and use it as a rather expensive portable drive.

Dropbox integrates nicely with your file manager, whether it’s Windows Explorer or Nautilus (Linux). You can simply drag your files into the designated Dropbox directory and the rest will sort itself out. It will simply upload the newly added file to the Dropbox servers. Some icons indicate the status of the upload, which means you’re well informed about the progress.

Files that were uploaded to the server are automatically downloaded when you fire up a computer that has Dropbox installed. The aforementioned icons also give you a nice indication of the status of the download.

When you sign up for Dropbox, you initially get 2GB of free storage space. That doesn’t sound like much these days, but I find that it’s usually more than enough. For some reason, you don’t store any files in a Dropbox account for any prolonged period, but rather for a few weeks or even just a few days.

If you really need more storage space in your Dropbox account, there are two options: either you pay or you try earning bonus space. You can earn extra space by inviting your friends to start using Dropbox through the referral program. If you’re willing to pay you can go pro. You’d pay US$9.99/month or US$99/year for 50GB of storage space. For 100GB you’d pay US$19.99/month or US$199/year.

Is going pro worth the extra cash? Hm, I must say I kind of doubt that. A small form factor external hard disk really isn’t that expensive any more and if you can do with no more than 16GB of storage space, a USB memory stick would be the way to go. Naturally, both the external hard disk and the USB memory stick carry the risk of being lost or stolen, which is not an issue with Dropbox. Remember, the files are always stored on your hard drive(s) after synchronization.

All in all, I myself am very satisfied with the basic service Dropbox offers. I have even earned a little bit of extra storage space by referring it to a friend. I hope to earn a bit more by getting more people to sign up based on my referral. So, if you’re interested in getting a Dropbox account, here’s the link to do so: sign up for Dropbox. I thank you for another 250MB of storage!