diff --git a/EcalAlCaRecoProducers/doc/alcareco_production.txt b/EcalAlCaRecoProducers/doc/alcareco_production.txt
new file mode 100644
index 00000000000..abafe0467f1
--- /dev/null
+++ b/EcalAlCaRecoProducers/doc/alcareco_production.txt
@@ -0,0 +1,10 @@
+/**
+ \page pAlcarecoProduction AlCaReco Production from AlCaRaw
+
+AlCaReco centrally produced have a bug, you need to run a reReco of the AlCaRaw with the Prompt GT: \ref pFindPromptGT
+
+Then follow the instructions for a rereco using as \b tag_name the name of the GT: \ref page5
+
+
+*/
+
diff --git a/EcalAlCaRecoProducers/doc/find_json.txt b/EcalAlCaRecoProducers/doc/find_json.txt
new file mode 100644
index 00000000000..6c75e0d2aa1
--- /dev/null
+++ b/EcalAlCaRecoProducers/doc/find_json.txt
@@ -0,0 +1,9 @@
+\**
+\page pFindJson Where to find the golden json
+
+The Golden Json is published by the certification group. Find the relative HN and subscribe if you are a regular user of ECALELF.
+
+The official json files are put in 
+\verbatim 
+/afs/cern.ch/cms/CAF/CMSCOMM/COMM_DQM/certification/
+\endverbatim
diff --git a/EcalAlCaRecoProducers/doc/find_prompt_gt.txt b/EcalAlCaRecoProducers/doc/find_prompt_gt.txt
new file mode 100644
index 00000000000..bbc3da22057
--- /dev/null
+++ b/EcalAlCaRecoProducers/doc/find_prompt_gt.txt
@@ -0,0 +1,13 @@
+/**
+\page pFindPromptGT How to find the Prompt GT
+
+Prompt GT is not any more in the file metadata. You need to access the DB to verify the GT used in Prompt recoing the dataset using DAS.
+
+You can do it from the command line using the following command:
+\code 
+das_client.py --query="config dataset=/DoubleEG/Run2015C-EcalUncalZElectron-PromptReco-v1/ALCARECO | grep config.global_tag" | tail -1
+\endcode
+
+
+*/
+
diff --git a/EcalAlCaRecoProducers/doc/introduction.txt b/EcalAlCaRecoProducers/doc/introduction.txt
new file mode 100644
index 00000000000..d2a57a89505
--- /dev/null
+++ b/EcalAlCaRecoProducers/doc/introduction.txt
@@ -0,0 +1,97 @@
+/**
+* \page page1 Introduction
+
+############################################################
+# ECALELF (Ecal CALibration with ELectrons Framework)  
+## study and calibration with Zee and E/p
+############################################################
+
+
+All the code needed to the ECAL studies and calibration with Zee is
+available in this package.
+
+The documentation of this packaged is provided in these pages extracted with doxygen from the source code and several txt files contained in the package itself.
+
+The code in this package takes the name of ECALELF 
+(Ecal CALibration with ELectrons Framework) 
+
+Informations for the implementation or re-implementation can be
+obtained looking the the various header files that will be completed
+with exaustive documentation (in the future)
+
+If you have doxygen, you can run the command
+\code doxygen fulldoc  \endcode
+from the Calibration/ directory and you can navigate instructions and code locally with
+\verbatim firefox doc/doxygen/fulldoc/html/index.html \endverbatim
+
+
+### The general calibration workflow is:
+\image html doc/ecalelf_schema.png
+\image latex doc/ecalelf_schema.eps
+
+ 1. start from AOD/RECO or AOD/RECO+RAW
+ 2. apply a ZSkim or a ZSCSkim or a WSkim selection to reduce the size of the dataset 
+ 3. produce the ALCARAW from AOD/RECO+RAW:
+    in this format most of the collections are
+    dropped and only the electrons, the uncalibrated rec hits
+    and few other collections like the beamspot, the conversions, the
+    rhoFastJet are kept
+
+ 4. Apply whatever ECAL tags (ADCtoGeV, IC, LC, Alpha) needed for test
+    or calibration and produce a new file in ALCARECO format.
+    The new file has a reduced collection of calibrated rechit of the
+    reco electrons, new SC collections produced with the new calibrated
+    recHits, the old reco electrons with a new SC associated
+    CAVEAT: all the electron variables are still the old RECO one,
+    only the reference to the new SC has been updated. This will not be
+    fixed, since it's a wanted feature that ensures that the selected
+    events are not changed and permit to check effectively the impact
+    of the calibration on the same events.
+
+ 5. calculate new electron energies with the EleNewEnergiesProducer
+    that creates ValueMaps that associate a float (the new energy) to
+    the electrons
+
+ 6. define if the electron pass some selections running the
+    EleSelectionProducer that creates valueMaps of floats with the
+    following meaning:
+    for cut-based selections 1=pass, 0=fail
+    for MVA selections the float is the output of the MVA
+
+ 7. for MC, produce an association map between reco electron and gen
+      electron (data-MC match)
+
+ 8. produce a patElectron collection where the electronID floats and
+    the new energies are imported by the patElectron and accessible by
+    electron.electronID("selectionName")
+    electron.userFloat("energyName") 
+    the gen electron identified by the data-MC matching step are
+    embended in the patElectron and accessible by:
+    electron.genLepton().eta()
+    electron.genLepton().energy()
+
+ 9. Filter the events requiring that the electron has passed at least
+    one selection
+
+ 10. dump the event content to a flat tree with one entry for Zee
+    candidate (the variables are array of two elements with one electron
+    per element)
+
+ 11. Run the analysis tools (Z fitter, smearing, etc.)
+
+ 
+
+In case the purpose is just to run on a particular dataset without any
+recalibration step, the 1-3a) steps can be replaced by
+
+ 1. Start from RECO or AOD
+ 2. Produce directly the ALCARECO, dropping all the unuseful
+     collections and reducing the recHit maps to the one related to
+     the electrons
+ 3. nothing 
+
+
+
+
+
+*/